Production and Characterization of Magnesium Oxychloride ... · production and characterization of...

PRODUCTION AND CHARACTERIZATION OF MAGNESIUM OXYCHLORIDE CEMENT BASED POLISHING BRICKS

FOR POLISHING OF CERAMIC TILES

A THESIS SUBMITTED TO THE GRADUATE SCHOOL OF NATURAL AND APPLIED SCIENCES

OF MIDDLE EAST TECHNICAL UNIVERSITY

BY

MUHAMMED SAİD ÖZER

IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR

THE DEGREE OF MASTER OF SCIENCE IN

METALLURGICAL AND MATERIALS ENGINEERING

DECEMBER 2008

Approval of the thesis:

PRODUCTION AND CHARACTERIZATION OF MAGNESIUM OXYCHLORIDE CEMENT BASED POLISHING BRICKS

FOR POLISHING OF CERAMIC TILES

submitted by MUHAMMED SAİD ÖZER in partial fulfillment of the requirements for the degree of Master of Science in Metallurgical and Materials Engineering Department, Middle East Technical University by, Prof. Dr. Canan Özgen _____________________ Dean, Graduate School of Natural and Applied Sciences Prof. Dr. Tayfur Öztürk _____________________ Head of Department, Metallurgical and Materials Engineering Prof. Dr. Abdullah Öztürk _____________________ Supervisor, Metallurgical and Materials Eng. Dept., METU Examining Committee Members: Prof. Dr. Muharrem Timuçin _____________________ Metallurgical and Materials Engineering Dept., METU Prof. Dr. Çetin Hoşten _____________________ Mining Engineering Dept., METU Prof. Dr. Abdullah Öztürk _____________________ Metallurgical and Materials Engineering Dept., METU Prof. Dr. Hakan Gür _____________________ Metallurgical and Materials Engineering Dept., METU Assist. Prof. Dr. Caner Durucan Metallurgical and Materials Engineering Dept., METU _____________________

Date: 15.12.2008

iii

I hereby declare that all information in this document has been obtained and presented in accordance with academic rules and ethical conduct. I also declare that, as required by these rules and conduct, I have fully cited and referenced all material and results that are not original to this work.

Name, Last name : Muhammed Said Özer

Signature :

iv

ABSTRACT

PRODUCTION AND CHARACTERIZATION OF MAGNESIUM

OXYCHLORIDE CEMENT BASED POLISHING BRICKS FOR

POLISHING OF CERAMIC TILES

Özer, Muhammed Said

M.S., Department of Metallurgical and Materials Engineering

Supervisor: Prof. Dr. Abdullah Öztürk

December 2008, 77 pages

Magnesium oxychloride cement (MOC) based grinding and polishing bricks

developed for polishing of granite based ceramic tiles were produced and

characterized. For surface grinding 46 and 180 grit size SiO2 powder embedded

MOC based abrasive bricks; for polishing 600 and 1200 grit size SiC powder

embedded MOC based abrasive bricks followed by neat (unreinforced) MOC pastes

were applied on ceramic tiles.

Three different neat MOC pastes depending on MgO/MgCl2 molar ratio in the paste

e.g. 6/1, 7/1, and 8/1, were formed and evaluated. Grinding bricks were formed by

adding 30 weight percentage, wt%, of both SiO2 powders. Polishing bricks were

formed by adding 20, 25, and 30 wt% of both SiC powders. X-Ray diffraction

analyses revealed that MOC F5 was the main crystalline phase in the neat MOC

pastes.

v

Additions of both SiO2 and SiC powders enhanced mechanical properties namely;

compressive strength and abrasion resistance, chemical durability in water and

polishing ability of MOC paste. More than 25 wt% addition of SiC powders had a

tendency to decrease the compressive strength and water resistance of MOC paste.

Polishing performance of abrasive bricks was evaluated in terms of mean surface

roughness of ceramic tiles and abrasive brick consumption upon polishing. Scanning

Electron Microscope examinations revealed the evidences of the reasons that 25 wt%

SiC powder embedded abrasive bricks has the best qualifications in terms of abrasion

resistance and polishing performance.

Keywords: Polishing, Granite, Ceramic Tile, Sorel Cement, Magnesium

Oxychloride, Abrasive, Roughness.

vi

ÖZ

SERAMİK KAROLARIN PARLATILMASINDA KULLANILACAK

MAGNEZYUM OKSİKLORİT ESASLI PARLATMA

PABUÇLARININ ÜRETİMİ VE KARAKTERİZASYONU

Özer, Muhammed Said

Yüksek Lisans, Metalurji ve Malzeme Mühendisliği Bölümü

Tez Yöneticisi: Prof. Dr. Abdullah Öztürk

Aralık 2008, 77 sayfa

Granit esaslı seramik karoların parlatılması için geliştirilmiş magnezyum oksiklorit

çimento (MOC) esaslı aşındırma ve parlatma pabuçlarının üretildi ve karakterize

edildi. Yüzey aşındırma işlemi için 46 ve 180 grit numaralı SiO2 tozu katkılı MOC

pabuçlar; parlatma işlemi için 600 ve 1200 grit numaralı SiC tozu katkılı MOC

pabuçları takiben yalın (katkısız) MOC pastaları seramik karoların üzerine sıra ile

uygulandı.

Çimento pastasının içerdiği MgO/MgCl2 mol oranı 6/1, 7/1 ve 8/1 olacak şekilde üç

ayrı MOC pastası hazırlandı ve değerlendirildi. Aşındırma pabuçları her bir SiO2

tozundan ağırlıkça % 30 oranı ilave edilerek hazırlandı. Parlatma pabuçları ise her bir

SiC tozundan çimentoya ağırlıkça % oranı 20, 25 ve 30 olacak şekilde ilave edilerek

hazırlandı. X-Işınları kırınım analizleri yalın MOC pastalarında kristal yapının esas

olarak F5 fazından oluştuğunu göstermiştir.

vii

SiO2 ve SiC tozlarının ilavesi MOC pastaların baskıda kırılma mukavemeti, aşınma

mukavemeti gibi mekanik özellikleri, suya karşı kimyasal dayanımı ve parlatma

performansını güçlendirmiştir. SiC tozunun %25’den daha fazla ilave edilmesi MOC

pastaların baskıda kırılma mukavemeti ve suya karşı kimyasal dayanımında

azalmaya sebep olduğu görülmüştür. Aşındırma pabuçlarının parlatma performansı

parlatma testi sonunda seramik karo yüzeyinin pürüzlülüğü ve harcanan aşındırıcı

miktarına bağlı olarak belirlendi. Seramik karo yüzeylerinde yapılan Taramalı

Elektron Mikroskobu incelemeleri ağırlıkça %25 oranda SiC tozu ilave edilen

pabuçların aşınma direnci ve parlatma performansı bakımından en üstün olduklarına

dair kanıtlar ortaya çıkarmıştır.

Anahtar Kelimeler: Parlatma, Granit, Seramik Karo, Sorel Çimento, Magnezyum

Oksiklorit, Aşındırcı, Pürüzlülük.

viii

To my

Mum & Dad

ix

ACKNOWLEDGMENTS

It is a great pleasure to thank my supervisor Prof. Dr. Abdullah Öztürk for his

scientific guidance, patient supervision and valuable constant encouragement at all

time. I have learned to think and understand in acedemic point of view from his

training. His support and tolerance helped me to accomplish my thesis.

I would want to acknowledge Prof. Dr. Muharrem Timuçin and each one of my

committee members for imparting their knowledge and motivating me to achieve this

goal.

I would like to thank to all of the staff of the Department of Metallurgical and

Materials Engineering for their contributions to my thesis studies. Especially, I want

to thank to my colleagues Aydın Rüşen, Metehan Erdoğan, Dr. Jongee Park for their

help and supports. I am also grateful to Assoc. Prof. Dr. Özgür Yaman for his advice

and guidance in conducting experimental studies of my thesis.

Present work was a part of SAN-TEZ poject entilted as “GRE Esaslı Seramik

Karoların Yüzeylerinin Parlatılmasında Kullanılabilecek Kompozit Parlatma

Disklerinin Üretimi”. The project was carried out together with Kaleseramik AŞ. and

supported by TUBITAK with the project number 105M140. I want to acknowledge

TUBITAK and Kaleseramik AŞ. for their financial and technical supports.

Lastly, I offer sincere thanks to my mum and dad for their continuous supports and

encouragements, great understanding and their endless love to achieve my goals at

every stage of my life. I thank to my brothers, sister, my wife and my dear doughter.

Without my family this study could not be accomplished.

x

TABLE OF CONTENTS

ABSTRACT...............................................................................................................iv

ÖZ...............................................................................................................................vi

DEDICATION.........................................................................................................viii

ACKNOWLEDGMENTS....................................................................................... .ix

TABLE OF CONTENTS..................................................................................... .....x

LIST OF TABLES...................................................................................................xii

LIST OF FIGURES................................................................................................xiii

CHAPTER

1. INTRODUCTION..................................................................................................1

2. THEORETICAL BACKROUND AND LITERATURE REVIEW..................4

2.1. CERAMIC TILES…………………………………………………….…4

2.1.1. Types of Ceramic Tiles..…………..……………………..... . ....4

2.1.2. Materials of Ceramic Tiles..……………………………….…..5

2.1.3. Sintering of Ceramic Tiles.......………………………….….…5

2.1.4. Structure and Properties of Ceramic Tiles......……………....…5

2.1.5. Applications of Ceramic Tiles..…………………………… .…6

2.2. POLISHING OF CERAMIC TILES.. ......…………........…………… ...6

2.3. MAGNESIUM OXYCHLORIDE CEMENT (MOC)....……………..…8

2.3.1. Materials and Formation of MOC............................................ ..8

2.3.2. Properties and Applications of MOC.......................…..…....... .9

2.4. ABASIVE BRICKS FOR CERAMIC TILE POLISHING.......…..…. ..10

2.5. PREVIOUS STUDIES............……………………………………… ...10

2.5.1. Previous studies on Ceramic Tiles................……………. …..11

2.5.2. Previous studies on MOC...................……………………… .12

3. EXPERIMENTAL PROCEDURE........…………....……....……………….…14

3.1. MATERIALS……………………………………………………….….14

3.2. PREPARATION OF ABASIVE BRICK SAMPLES......…………….. 16

3.2.1. Equipments.....……………………………………………… .16

3.2.2. Moulds……………………………………………............. …16

xi

3.2.3. Mixing and Molding Procedure..........……………………. …16

3.2.4. Compositions of Abrasive Samples.........................………....17

3.2.5. Storing of hte Samples.......................................................... . ..18

3.3. TESTING METHODS AND EQUIPMENTS........................................18

3.3.1. Characterization............................................ .......... ................18

3.3.2. Density Measurements ......................................................... .19

3.3.3. Compression Tests......................................................... ..........19

3.3.4. Abrasion Tests..........................................................................19

3.3.5. Water Resistance................................................................. .....20

3.3.6. Polishing Tests...................................................... ................. .21

3.3.6.1. Polishing Test Samples..... ........................................21

3.3.6.2. Experimental Setup of Polishing Tests.................... .21

3.3.6.3. Polishing Test Parameters....................................... .22

3.3.6.4. Testing Procedure................................................... .23

3.3.6.5. Polishing Performance............................................ .24

3.3.6.5.1. Abrasive consumption................................24

3.3.6.5.2. Surface Roughness .. ............................... .24

4. RESULTS AND DISCUSSION…………………………………..…. ………...26

4.1. CHARACTERIZATION…………………….... ……..........… ………26

4.2. DENSITY MEASUREMENTS.................................................... ...…...32

4.3. COMPRESSION TESTS..........…………………….…........….….…...35

4.4. ABRASION TESTS....................... ………………….….........……. ....39

4.5. WATER RESISTANCE...................................................................... . ..44

4.6. POLISHING TESTS.............................................................................. .49

4.7. MICROSTRUCTURAL EXAMINATIONS..........................................56

4.7.1. Neat Sorel Cement Microstructures........................................ .56

4.7.2. SEM Micrographs of Polished Ceramic Tiles....……........... . .60

5.CONCLUSIONS................................................................................................. ..67

REFERENCES.........................................................................................................69

xii

LIST OF TABLES

Table 3.1. Chemical composition of MgO powder in wt% ................................. ...14

Table 3.2. Chemical composition of MgCl2.6H2O salt in wt%..............……. …......14

Table 3.3. Chemical composition of SiO2 powders in wt%...............…........ ...........15

Table 3.4. Chemical composition of SiC powders in wt% . .................................. ..15

Table 3.5. Physical properties of SiC powders ……………..............…....... ............15

Table 3.6. Molar composition of neat MOC and abrasive powder content of the

bricks used in this study .................................................................... ……17

Table 4.1. Particle size distribution analyses of raw materials. .............................. .28

xiii

LIST OF FIGURES

Figure 2.1. Schematic represantation of tangential head polishing.. ......................... .8

Figure 3.1. Schematic drawing of polishing test equipment................................... . .22

Figure 3.2. Flowchart showing the experimental procedure of abrasive brick

preparation....................................................................................................25

Figure 4.1. Particle size distribution of 46 grit SiO2 powder…………………… …27

Figure 4.2. Particle size distribution of 180 grit SiO2 powder…………………… ..27

Figure 4.3. Particle size distribution of 600 grit SiC powder……………………... .27

Figure 4.4. Particle size distribution of 1200 grit SiC powder…… ..………………28

Figure 4.5. Particle size distribution of MgO powder……………… ...……………28

Figure 4.6. X-Ray diffraction pattern of MgO powder………………… ………….29

Figure 4.7. X-Ray diffraction pattern of neat MOC pastes…….…………. ….……30

Figure 4.8. X-Ray diffraction pattern of neat MOC pastes…….……………. .……31

Figure 4.9. Variation in density of the neat MOC pastes with MgO/MgCl2

molar ratio of the paste…………………… ..……………………………...32

Figure 4.10. Density of different mixtures investigated in this study……. ………..33

Figure 4.11. Variation in density of 600 and 1200 grit size SiC powder

embedded abrasive bricks with SiC powder addition……………. ………34

Figure 4.12. Variation in compressive strength of the MOC pastes with

MgO/MgCl2 ratio in the paste…… ........................................... …………36

Figure 4.13. Compressive strength of different mixtures investigated in this study..37

Figure 4.14. Variations in compressive strength of abrasive bricks with

600 and 1200 grit size SiC powder additions……………… .…………….38

Figure 4.15. Variation in weight loss and height loss occurred during abrasion

test of the neat MOC pastes with MgO/MgCl2 molar ratio in the paste.. …40

Figure 4.16. Variation in weight loss and height loss occurred during

abrasion tests with SiO2 powder addition to neat MOC……... ..…………..41

Figure 4.17. Variation of weight loss and height loss occurred during

abrasion tests of the 600 grit size SiC powder embedded abrasive

bricks with SiC powder addition. …………………………………………42

xiv

Figure 4.18. Variation of weight loss and height loss occurred during

abrasion tests of the 1200 grit size SiC powder embedded

abrasive bricks with SiC powder addition…………………………………42

Figure 4.19. Variation in water absorption and weight loss after immersion

of the neat MOC pastes with MgO/MgCl2 molar ratio in the pastes… . ….44

Figure 4.20. Water absorption and weight loss of different mixtures

investigated in this study…......................................................................45

Figure 4.21. Variation in water absorption and weight loss occurred during

immersion tests of the 600 grit size SiC powder embedded abrasive

bricks with SiC powder addition…......................................................... ..46


immersion tests of the 1200 grit size SiC powder embedded abrasive

bricks with SiC powder addition........................................................... …47

Figure 4.23. Mean surface roughness of ceramic tiles and abrasive

consumption occurred during the application of 7-Q-46, 7-Q-180

and 7-25-C coded abrasive bricks… .................................................... …49

Figure 4.24. Surface roughness profile of ceramic tile after the application

of abrasive bricks, a) 7-Q-46 and b) 7-Q-180….................................... …49

Figure 4.25. Mean surface roughness of ceramic tiles after being exposed

to different polishing procedures with regard to grit size of

the abrasive embedded…........................................................................ .50

Figure 4.26. Sample roughness profile of ceramic tile after polishing

with, a) 7-25-C and b) 7-25-F …. ........................................................... .51

Figure 4.27. Variations in abrasive brick consumption with grit size of

SiC powder for different amounts of SiC powder addition…................... .52

Figure 4.28. Variation in surface roughness of ceramic tile and abrasive

brick consumption upon 1600 grade polishing practice with

MgO/MgCl2 molar ratio in the paste................................................... ….54

Figure 4.29. Surface roughness profile of ceramic tile after being polished

with paste, a) 6-00, b) 7-00 and c) 8-00 ................................................ …55

Figure 4.30. SEM fractographs of the neat MOC, a) paste 6-00, b) paste 7-00

and c) paste 8-00…................................................................................. .57

xv

Figure 4.31. SEM micrographs of the needle shaped MOC crystals in,

a) paste 6-00, b) paste 7-00 and c) paste 8-00…........................................ .59

Figure 4.32. SEM micrographs of ceramic tile surface after being exposed

to 600 grit polishing practice with abrasive bricks of, a) 7-20-C,

b) 7-25-C and c) 7-30-C. ..................................................................... …62

Figure 4.33. SEM micrographs taken from polished surface of ceramic tile

after being exposed to 1200 grit polishing practice with abrasive

bricks of a) 7-20-F, b) 7-25-F and c) 7-30-F… ...................................... . .64

Figure 4.34. SEM micrographs taken from polished surface of ceramic tile

after being exposed to the neat MOC of a) paste 6-00, b) paste 7-00

and c) paste 8-00… ................................................................................. .66

CHAPTER 1

INTRODUCTION

Polished ceramic tiles have attracted great interest in recent years due to their

superior properties in chemical, mechanical and aesthetic aspect [1-8]. Ceramic tile

body consists of low amount of mullite and high amount of residual quartz as

crystalline phases which are dispersed in the feldspatic glassy matrix [6, 9-11]. High

glass content (> 40% by volume) and low porosity (< 0.5% by volume) are typical

properties of ceramic tiles that enable them for surface treatment [6, 7, 9, 11].

Surface polishing is an expensive and critical step in production of ceramic tiles [1-

3, 5, 11]. Aesthetic properties of the ceramic tiles in terms of gloss and roughness

are improved so much that any method but only polishing process can achieve this

level [1, 2, 6]. However, opening of the closed pores already existing in the ceramic

body is an unavoidable drawback of surface polishing. Moreover, if process

parameters are not applied properly surface quality of the ceramic tile decreases due

to scratches, grooves, cuts and subsurface cracking and polishing cost increases due

to high amount of abrasive consumption [1, 3, 6, 10].

Surface polishing of ceramic tiles is achieved in three main steps; initial leveling,

grinding and fine polishing. The first step involves flattening of the tile surface by

diamond abrasive tools. Second is the preparation step of the tile surface for

polishing by reducing roughness. It usually consists of 8 grades (46-60-80-120-180-

240-320-400) with respect to the grit number of quartz and SiC particles in the

abrasive bricks. Third step is the actual polishing operation where surface gloss of

2

the ceramic tile is improved. In this step, 600-800-1000-1200-1400-1500-1600 grit

abrasive bricks are used successively.

Most common abrasive bricks for fine polishing of ceramic tiles are magnesium

oxychloride cement (MOC), also known as Sorel cement, bonded SiC composite

bricks [3, 11-15]. MOC is used as a binder in the production of polishing bricks due

to its high early strength, excellent abrasion resistance and good bonding ability to

aggregates [14-19].

MOC is a type of hydraulic cement formed by mixing proper ratio of caustic

calcined MgO powder and MgCl2 solution [15-28]. Four major crystalline phases

are defined for MOC according to molar ratio of Mg(OH)2 to MgCl2 in the chemical

composition: 2Mg(OH)2.MgCl2.4H2O (F2), 3Mg(OH)2.MgCl2.8H2O (F3),

5Mg(OH)2.MgCl2.8H2O (F5), and 9Mg(OH)2.MgCl2.H2O (F9) [15-19, 24-27].

However, room temperature stable phases are only F3 and F5 among which the

latter is reported to have better properties in terms of chemical durability and

strength [13, 16, 18, 19, 28, 29].

High concentrations of MgO (MgO/MgCl2 greater than 5) in the cement paste

favors formation of F5 phase, on the other hand, low MgO concentrations leads to

the formation of F3 phase [16-19, 28-30]. In addition to MgO/MgCl2 molar ratio,

MgO/H2O ratio is an important aspect for mechanical and chemical properties. Low

water content is favorable for strength and durability of the pastes. However; the

minimum amount of water is limited by workability of the paste [15, 16]. Thus,

MOC pastes with high MgO and low water content are desirable [13, 15, 18].

In spite of some investigations [3, 6, 10-19] on the formation, properties and

structure of MOC and ceramic tile polishing, data on the polishing performance of

abrasion and polishing bricks with respect to their chemical compositions are rare

and sparse. It is necessary to conduct a research to determine and improve the

polishing performance of the polishing bricks if they are to be technologically

useful. Otherwise the bricks may fail during service and/or may cause the formation

3

of nicks, scratches, cracks etc. on the surface of the tiles. An understanding of the

polishing performance of these bricks is important also for correlating the abrasion

and polishing behavior with the type of abrasives used in the system. Hence, studies

on polishing bricks have both scientific and practical significance.

Present study was undertaken to determine the factors affecting the production and

polishing performance of fine polishing abrasive bricks and to manipulate these

factors to upgrade the product quality. Effects of molar composition of binder

phase, type (SiO2 or SiC), and amount of abrasive phase on the abrasion and

polishing performance were clarified and discussed in detail. H2O /MgCl2 ratio of

11/1 was held constant for all cement pastes. MgO/MgCl2 molar ratio of neat

cement phase was studied between the range of 6/1 – 8/1. MgCl2/MgO ratio of 7/1

was kept constant for the production of SiO2 and SiC powder added composite

abrasive bricks. Two grades for grinding step (46 and 180) and three grades of

polishing step (600, 1200, and 1600) were chosen for experiments. For the coarse

grinding (abrasion), 46 and 180 grit size quartz powder embedded MOC bricks

were used. For 600 and 1200 grades of polishing, 600 and 1200 grit size SiC

powder embedded MOC bricks were used. Neat (unreinforced) MOC samples were

used for 1600 grade polishing.

Particle size distribution of quartz, SiC, and MgO powders were analyzed.

Crystalline phases present in the MgO powder and in the neat cement were

identified by X-Ray diffraction technique. Single axis compression test and

abrasion test were employed for measuring mechanical properties. Chemical

durability against aqueous media (water resistance) was determined via water

absorption and weight loss of samples after immersing into de-ionized water.

Microstructure development in cement matrix and surface properties of abraded and

polished ceramic tiles were investigated by scanning electron microscopy. Polishing

performance of polishing bricks was evaluated in terms of surface roughness of the

polished ceramic tile together with weight loss of the abrasive brick occurred during

polishing test.

4

CHAPTER 2

THEORETICAL BACKROUND AND LITERATURE REVIEW

In this chapter, initially introductory information about ceramic tiles was given.

Secondly, polishing process of ceramic tiles was described. Then, features of

abrasive bricks and magnesium oxychloride cement (MOC) were explained in

detail. Finally, previous studies concerning the subjects were summarized. Various

testing methods that were applied to ceramic tiles and cement based materials by

previous investigators were also mentioned in the last part.

2.1. CERAMIC TILES

2.1.1. Types of Ceramic Tiles

Ceramic tiles are divided into two main groups as glazed and unglazed ceramic

tiles. Glazed ceramic tiles, differing from the unglazed ceramic tiles, include

prolonged firing of the ceramic body and glazing application. Furthermore, firing

procedure and composition of glazed tiles are so adjusted that a highly crystalline

body and a glassy glaze is achieved. On the other hand, unglazed ceramic tiles, also

called as porcelain tiles, are produced through a fast-single firing operation and

includes high amount of glassy phase which enables the tiles for surface polishing

operations [4-9, 13].

5

2.1.2. Materials Used in the Production of Ceramic Tiles

Raw materials of ceramic tiles are mainly composed of clay, feldspar and quartz.

Clay provides plasticity and is considered as a source of Al2O3. Feldspar is used as

fluxing agent and quartz plays role for viscosity of the body during sintering and

contributes to mullite formation [7]. A general compositional range for the

components are 40-50 clay (kaolin), 35-45 feldspar, and 10-15 quartz by weight

percentage, wt% [4].

2.1.3. Sintering of Ceramic Tiles

After mixing, batching, and pressing of raw materials, ceramic tiles are subjected to

high temperature for sintering purpose. Developments in materials and production

techniques of ceramic tile industry yield a single firing method for ceramic tile

processing [31-33]. Main idea of this process is both reducing energy costs of

sintering and improving properties of product. Raw materials composition and

properties are prepared in order to get a glassy matrix and mullite crystal formation

at elevated temperatures such as 1200 ºC within an hour [4, 9, 13].

Firing duration cold-to-cold is usually applied about 1 h. Because of the high

temperature and short firing period, ceramic tile sintering process is described as a

non-equilibrium process. Dissolution of chemical components, formation of mullite

crystals and glassy matrix development are kinetically controlled processes [4].

2.1.4. Structure and Properties of Ceramic Tiles

Chemical, mechanical and aesthetic properties of ceramic tiles are eminent [1-7].

They are composed of high amount of glassy phase, in-situ formed mullite crystals

and residual quartz phase. Mullite crystals are generally reported as 7-8 wt% of the

body, glassy matrix constitutes more than 40 wt% and the rest is typically

composed of quartz. A vital property of ceramic tiles is the low amount of porosity

6

which is usually below 0.5 volumetric percentage, vol% [6-11]. Low amount of

porosity and high amount of glassy phase make the tiles feasible for surface

treatment (polishing) processes. The polishing process multiples the aesthetic

appearance of the material with adequate gloss and low roughness.

2.1.5. Applications of Ceramic Tiles

Ceramic tiles are commonly used covering materials for both indoor and outdoor

applications of buildings and houses [12, 13]. Due to high gloss and smooth

surfaces they exhibit easy cleaning and aesthetic appearance [1, 3, 4, 6, 11]. The

interest to both as-fired and polished ceramic tiles is growing in recent years.

2.2. POLISHING OF CERAMIC TILES

For improving ceramic tiles surface properties in terms of smoothness (roughness)

and gloss, they are subjected to surface polishing operation after sintering process

[3, 6, 10-13]. Despite the beneficial effects on appearance of ceramic tiles, polishing

process exhibits difficulties and drawbacks such as opening of closed pores,

grooves and high operational costs [1-3, 10].

Surface polishing of ceramic tiles are achieved on long polishing lines from coarse

grade to fine grade steps. Actually, the whole surface treatment can be classified in

three parts:

In the first part, diamond discs are used for leveling of the tiles. Ceramic tile surface

is highly damaged and deeply scratched. Most of the material loss from tile surface

occurs on this step. The diamond discs are usually called as long-life materials to

express they are used for prolonged durations [10-13].

7

The second step can be called as grinding operation. In order to clear deep traces of

diamond discs, coarse grit size abrasive powder embedded abrasives are used in

grinding step. Starting from 36 or 46 grit size toward 400 grit size abrasives are

used successively in 8 to 9 steps (36-46-60-80-120-180-240-320-400). In this step,

ceramic tile is prepared for the polishing step. The roughness of ceramic tile

decreases drastically [3, 11].

The third step, which is the main scope of this study, includes final polishing of

ceramic tile. In this step, abrasive bricks are generally produced with MOC binder

and very fine SiC powder as abrasive addition to the binder [11-15]. Grit sizes of

abrasives are generally in the order of 600-800-1000-1200-1400-1500-1600 in the

polishing line. Excellent gloss and smoothness of tile is gained in this final

polishing step. For final quality of product, this step has crucial importance not to

leave any process damage on the tile surface [3, 11-15].

Ceramic tile polishing system has various operational methods for different

purposes. These methods are mainly classified as “Tangential Head”, “Planetary

Head” and “Fixed Head” systems [3]. However, second and third step of surface

polishing are mostly applied with “Tangential Head System”. Tangential Head

System includes simultaneous motions of abrasive bricks in double axis. The group

of 6 to 8 abrasive bricks is mounted on a circular flange which rotates around the

central axis of itself. Meanwhile the rotation of flange, every abrasive brick

oscillates with a 15º on the horizontal axis of abrasive brick itself. By this way, an

oscillating tangential line of the abrasive brick makes contact with the ceramic tile

during polishing. Hutching et al. [3] reported a schematic representation of

“Tangential Head” polishing which is given in Figure 2.1.

8

Figure 2.1. Schematic represantation of tangential head polishing. (a) A rotating

head carries six abrasive blocks and the tiles are transported slowly beneath the

head, (b) detail of a single abrasive block, which oscillates in a small arc with radius

R about horizontal axis while the whole polishing head rotates about vertical axis.

2.3. MAGNESIUM OXYCHLORIDE CEMENT (MOC)

2.3.1. Materials and Formation of MOC

The MOC, also known as Sorel cement, pastes are mixtures of MgO powder and

MgCl2 solution in proper compositions [15-28]. Four major crystalline phases exist,

defined with respect to molar ratio of Mg(OH)2/MgCl2 in the chemical composition

of cement: 2Mg(OH)2.MgCl2.4H2O (F2), 3Mg(OH)2.MgCl2.8H2O (F3),

5Mg(OH)2.MgCl2.8H2O (F5), and 9Mg(OH)2.MgCl2.H2O (F9) [15-19, 24-27].

9

However, room temperature stable phases are only F3 and F5 among which the

latter is reported to have better properties in terms of chemical durability and

strength [13, 16, 18, 19, 28, 29].

High concentrations of MgO (MgO/MgCl2 ratio greater than 5) in the cement paste

favors formation of F5 phase, on the other hand, low MgO concentrations leads to

formation of F3 phase [16-19, 28-30]. In addition to MgO/MgCl2 molar ratio,

MgO/H2O ratio is an important aspect for mechanical and chemical properties. Low

water content of the pastes is favorable for strength and durability. However, the

amount of water is limited by workability of the paste [15, 16]. Thus, for production

of polishing bricks, the MOC pastes with high MgO and low water content are

desirable [13, 15, 18].

2.3.2. Properties and Applications of MOC

The MOC pastes have low thermal conductivity and good fire resistance [17]. Other

than aesthetic appearance of MOC, they exhibit high compressive strength, abrasion

resistance and good bonding strength to second phases like sand or gravel [13].

Due to its high performance in abrasion resistance and strength, the MOC pastes

find use in abrasion brick industry [3, 13, 15]. Other uses of MOC are reported as

restoration and decorative purposes [17], flooring materials in industry [24], and

treatment of industrial wastes [34].

Failure of MOC due to low chemical durability against atmospheric CO2 and

moisture has been reported. Additionally, lack of resistance to weathering and

leaching out of MgCl2 from the cement limits the use of MOC [15, 17, 35].

10

2.4. ABRASIVE BRICKS FOR CERAMIC TILE POLISHING

Various types of abrasive bricks depending on type of binder material and abrasive

powder were developed for polishing of ceramic tiles. Commonly used binder

materials for abrasive brick applications are epoxy resin and magnesium

oxychloride cement [3, 6, 10-15]. Abrasive powder choice on the other hand,

depends on the quality and purpose of bricks. Namely, for coarse grinding purposes

quartz and SiC particles are used together as abrasive ingredient. Pumice type

porous materials are also reported to be used for more economic grinding bricks

especially for marble. However, third step polishing bricks for ceramic polishing

are mostly produced with only SiC powder addition to MOC [13, 14, 18].

2.5. PREVIOUS STUDIES

2.5.1. Previous Studies on Ceramic Tiles

Production procedures of unglazed ceramic tile techniques are recently developed

and still being improved [1-10]. There are two important echoes of unglazed

ceramic tiles to ceramic industry: reduction of production cost and improvement of

product properties. Consequently, researches on the subject cover mostly material

and process costs. Compositional parameters are studied to obtain possible

opportunities of waste material utilization in ceramic tile production [2, 7, 36-38].

In addition to composition, effect of firing temperature is also a subject of interest.

Researchers studied a certain composition of a ceramic tile body for estimating a

proper firing temperature [4]. A range of 1260-1280 ºC has been stated as an

optimum temperature range in terms of the amount of open and/or closed porosity,

crystallization behavior and mechanical properties of the ceramic body.

11

Romero et al. [9] studied kinetic parameters of mullite phase formation in ceramic

tiles with Differantial Thermal Analysis (DTA) and reported crystallization

temperature and activation energy values for a certain ceramic tile composition.

Microstructural studies have been conducted to correlate the compositional and

textural properties of ceramic tiles with surface polishability [6, 10]. The

investigations revealed the implications of quartz content, initial size of quartz

powder and firing temperature of the ceramic tile on development of porosities.

Not only material property of ceramic tiles, but also polishing process was studied

by various researchers [3, 6, 10-13]. Hutchings et al. [3] proposed a simple

polishing test setup for polishing tests of ceramic tiles. The proposed system

resembles to metallographic polishing method. Ceramic tile to be polished is fixed

on a rotating disc and abrasive sample is held on the rotating tile. In the meantime,

the abrasive sample is rotated on its vertical axis too. Consequently, double

dimensional abrasion action is achieved. For the experimental polishing studies of

this publication, the polishing setup designed by Hutchings et al. [3] was adopted.

Sarı et al. [39] studied polishability of various kinds of marble surfaces and defined

a quantitative approach to gloss measurements of Turkish marbles. They have

related the glossiness of the material with both structural properties and roughness

degree. In another study on marble polishing [40], a three dimensional testing

method for surface analyses was proposed. It is a kind of surface mapping system

that could be used for measuring surface roughness of marble or granite tiles. The

roughness range studied was mostly for coarse grinding levels such as 40 to 50 µm.

Polishing of ceramic tiles was considered as not only aesthetic but also safety

aspects in various studies. Chang [41] reported the importance of friction

phenomena and mentioned about occupational injuries resulting from slips and

falls. A correlation between surface properties of ceramic tiles (roughness),

contaminant material (water or oil) and the slipping surface (rubber sheet of shoe

bottom) has been established.

12

2.5.2. Previous Studies on MOC

MOC or Sorel cement history goes back to 19th century when “S. T. Sorel” first

invented the material which is named with the author’s name in honor of him [25].

Since then, several researches have been conducted on the subject to understand and

improve the properties of MOC. The first attempts identified ternary MgO-HCl-

H2O system and crystalline phase development of MOC [21]. Further researches

have revealed F5 and F3 are thermodynamically and chemically stable phases at

room temperature [24-27].

Dehua and Chuanmei [29] studied phase formation mechanisms in the MOC pastes

and strongly stressed that crystallization of cement phase does not occur directly

upon MgO and MgCl2 compounds. A primary step has been described for

dissolution of ionic species of Mg++, Cl-, OH-, and H+ in cement paste upon mixing

the components. The first step was complexation of those ions into complex specie

forms. They mentioned a medium gel formation and bridging reactions of complex

species with the ligand H2O and Cl- ions. Crystallization of cement phase occurs as

the result of those final bridging reactions. Studies on crystalline phases of MOC

revealed that low MgO concentration in cement paste yields F3 phase formation,

but; if MgO happens to exist in large amounts (MgO/MgCl2 molar ratio to be more

than 5/1) in the cement paste F5 phase formation is favored [18, 19, 24-26].

Microstructural examinations by Bury et al. [22] and by Dehua et al. [29] revealed

that crystalline phases F3 and F5 of MOC are in elongated (needle shape) crystals.

Performance and properties of MOC were investigated by various researchers.

Different compositions with different raw materials were reported for preparation of

MOC pastes. Harper [23] studied the effect of calcination temperature of the

magnesite for production of MOC and emphasized the importance of activity and

surface area of MgO powder on the reaction kinetics and resulting mechanical

properties of the cement paste. Major factors that affect properties of MOC pastes

are listed as, calcination temperature and duration, powder size, porosity, surface

13

area and purity of MgO powder. Additionally, ratios of MgO/MgCl2 and MgO/H2O

in the paste are major determinant compositional factors for the properties of MOC.

Li et al. [16] studied the effects of composition on MOC and highlighted the

importance of reactivity of MgO powder. They disclosed that MgO/MgCl2 molar

ratio is the most critical design parameter. The mechanical tests were supplemented

with X-Ray diffraction (XRD) analyses. They anticipated optimum MgO/MgCl2

and H2O/MgCl2 ratios as 13 and 12, respectively for their choice of raw materials.

MOC was used as a waste treating material [42]. Refuse materials especially

inorganic construction wastes were embedded in MOC to be used for outdoor floor

tile production. Environmentally beneficial aspects of using urban refuse as filler

material for a new product were stated.

Junsong [31] reported a foamed type material that is prepared with MOC.

He mentioned the effect of phosphoric acid addition into MOC as a foaming agent.

A small amount of phosphoric acid is said to be foaming the structure and covering

cement crystals forming a protective layer against aqueous media. The XRD

analyses revealed that cement phase formation is not affected with the addition.

Castellar et al. [15] worked on a particular case of MOC based polishing bricks.

The samples of the study are reported to exhibit cracks during storage in depots.

The reason of cracking was recognized as structural stress occurring due to

carbonation reactions. Similar approach was applied for a case study which was

about marble restorating MOC mortar [17]. The study includes XRD analyses of

outdoor cement materials. The carbonation of MOC due to CO2 gas and humidity in

the atmosphere was underlined by the authors.

Present experimental study was conducted with the light of previous studies on

MOC pastes, ceramic tiles, and polishing process. Understanding of production

parameters for MOC based abrasive bricks and polishing process was ascertained.

14

CHAPTER 3

EXPERIMENTAL PROCEDURE

In this chapter, first the characteristics of the materials used for the production of

magnesium oxychloride cement (MOC) based polishing bricks were given. Then,

the sample preparation procedure was explained. Finally, testing methods and

equipment employed in the experimental studies were described in detail.

3.1. MATERIALS

For preparation of the MOC pastes, caustic calcined MgO powder obtained from

“Grecian Magnesite Co., Greece” was used. Chemical composition of MgO powder

in weight percentage, wt%, is given in Table 3.1.

Table 3.1. Chemical composition of MgO powder in wt% [43].

MgO SiO2 CaO Fe2O3

96.0 1.5 2.0 0.5

MgCl2.6H2O salt was obtained from “Nedmag Co., Netherland” in pellet form.

Chemical composition of MgCl2.6H2O in wt% is presented in Table 3.2.

Table 3.2. Chemical composition of MgCl2.6H2O in wt% [44].

MgCl2 CaCl2 Fe2O3 H2O

47.0 2.0 0.2 50.8

15

Two grades of SiO2 powders supplied from “Akyol TAŞ.” were used for the

preparation of coarse grinding abrasive bricks. Particle size of the SiO2 powders

corresponds to the grit numbers of 46 and 180. Chemical composition of the SiO2

powders in wt% is given in Table 3.3.

Table 3.3. Chemical composition of SiO2 powders in wt% [45].

SiO2 Al2O3 Fe2O3 K2O

> 98.0 < 0.6 < 1.0 < 0.4

Two grades of SiC powders supplied from Eczacıbaşı Esan AŞ. were used for

preparation of composite abrasive bricks. Particle size of the SiC powders

corresponds to the grit numbers of 600 and 1200. Chemical compositions of the SiC

powders in wt% are given in Table 3.4.

Table 3.4. Chemical composition of the SiC powders in wt% [46].

SiC powder size SiC Fe2O3 C (Active)

600 grit > 97.0 < 1.3 < 0.3

1200 grit > 96.0 < 1.4 < 0.3

In addition to chemical composition, physical properties of the SiC powders given

by company specifications are listed in Table 3.5.

Table 3.5. Physical properties of SiC powders [46].

Mineral Type Crystal System Crystal Color Mohs Hardness

Alfa-SiC Hexagonal Black 9.15

16

3.2. PREPARATION OF ABRASIVE BRICKS

3.2.1. Equipment

Raw materials to prepare neat cement pastes and SiO2 or SiC powders added

abrasive bricks were weighed accurately (±1 g) using a Beurer KS30 model kitchen

scale which had 3 kg capacity and 1 g precision.

A KitchenAid trademark “Tilt-Head Stand Mixer” with 3 lt capacity was used for

mixing of the paste mixtures. The mixer had ten level of operating speed and double

axis rotating paddle (KitchenAid “Flat Beater”) for improved mixedness.

3.2.2. Moulds

Two sorts of home-made plastic moulds, one in cylindrical the other one in cubic

shape, were used. The cylindrical moulds had a cross-section area of 8 cm2 and

height of 4 cm. Nominal dimensions of the cubic molds was 7x7x7 cm.

3.2.3. Mixing and Molding Procedure

Mixing of pastes was achieved simply in three steps: In the first step, MgCl2.6H2O

salt of required amount to obtain the molar composition of the desired paste was

dissolved in water. Molar ratio of H2O/MgCl2 in the cement pastes was kept

constant with 11/1 for all type of neat and composite samples. In view of the fact

that there are 6 moles of chemically bonded H2O in the MgCl2.6H2O salt, 5 moles

of extra water was added for every moles of MgCl2. Salt-water mixture was blended

for 10 min to dissolve Mg++ and Cl- ions in the aqueous media. In the second step,

the necessary amount of MgO powder was added into MgCl2 solution. For SiO2 or

SiC powder embedded abrasive bricks, proper amount of SiO2 or SiC powder was

17

admixed to the paste. All pastes were mixed for 30 min to obtain homogeneity and

uniformity of the materials. After mixing operation, pastes were cast into moulds.

3.2.4. Compositions of Abrasive Bricks

Three different neat MOC pastes were prepared with regard to MgO/MgCl2 molar

ratio (6/1, 7/1 and 8/1) in the cement. The ratio of 5/1 and 9/1 were not used for the

tests because the former lacked in setting reactions. The latter cracked and broken

down into pieces for about 2 days after setting probably due to high amount of

internal stress [13]. Molar composition of neat MOC and abrasive powder content

of the bricks used in this study were presented in Table 3.6. Hereafter the samples

will be referred to with the codes given in Table 3.6.

Table 3.6. Molar composition of neat MOC and abrasive powder content of the

bricks used throughout this study.

Sample

Code

Proportions of raw materials

(MgO:MgCl2:H2O molar ratio)

Additive

grit size

SiC content

(wt%)

SiO2 content

(wt%)

6-00 6 MgO : 1 MgCl2 : 11 H2O none none none



7-Q-46 7 MgO : 1 MgCl2 : 11 H2O 46 none none

7-Q-180 7 MgO : 1 MgCl2 : 11 H2O 180 none none

7-20- C 7 MgO : 1 MgCl2 : 11 H2O 600 20 none

7-25- C 7 MgO : 1 MgCl2 : 11 H2O 600 25 none

7-30- C 7 MgO : 1 MgCl2 : 11 H2O 600 30 none

7-20- F 7 MgO : 1 MgCl2 : 11 H2O 1200 20 none

7-25- F 7 MgO : 1 MgCl2 : 11 H2O 1200 25 none

7-30- F 7 MgO : 1 MgCl2 :11 H2O 1200 30 none

18

Molar ratio of MgO/MgCl2 was 7/1 for all types of abrasive powder admixed

abrasive bricks. SiO2 powder was added only in the amount of 30 wt% for both 46

and 180 grit size. SiC powder addition was made in the amounts of 20, 25, and 30

wt% for both 600 and 1200 grit size. SiC powder embedded bricks were used in 600

and 1200 grade polishing tests. SiC addition of over 30% was not applied due to

workability problems of the cement pastes.

3.2.5. Conditioning of the Samples

Neat MOC and abrasive powder embedded abrasive bricks were kept 28 days in the

plastic moulds in accord with TS 1769 standard [47]. Relative humidity of

conditioning room was about 50% and temperature was in the range of 20-25 ºC.

After 28 days of conditioning period samples were de-molded and tested.

3.3. TESTING METHODS AND EQUIPMENT

3.3.1. Characterization

Particle size analyses of SiO2, SiC and MgO powders were conducted using a

Malvern Mastersizer 2000 model particle size analyzer. Average of three

measurements for each powder sample was taken to get reliable data for particle

size distribution analyses.

X-Ray diffraction (XRD) technique was applied to identify the crystalline phases

present in the MgO powder and MOC pastes. Rigaku Geigerflex (DMAX/B) model

X-Ray diffractometer was employed for analyses. Samples were scanned between

10º and 65º with a speed of 2 º/min using Cu-Kα radiation. Microstructure

examinations of neat cement pastes and polished ceramic tile surfaces were

performed by Jeol 6400 model Scanning Electron Microscope.

19

3.3.2. Density Determination

In determining density of the samples, the cylindrical and cubic shape samples were

weighed by 1 mg precision. Density of the samples was computed by dividing the

mass of the sample to volume of the sample. At least ten samples were tested for

each brick to get a reliable data on density determinations. Average value of the ten

determinations was calculated in g/cm3 and taken as density of the brick.

3.3.3. Compression Test

Compression tests were performed using a compression test machine having a

capacity of 300 tons (Baz Makine AŞ.). The tests were conducted at a loading speed

of 240 kgf/s. Six samples in nominal dimensions of 7x7x7 cm were tested for each

paste to determine the compressive strength of the bricks. Fracture load of the

samples were recorded. Compressive strengths of the bricks were calculated using

the formula

σ = F/A

where; σ represents compressive stress, F is the maximum force applied, and A is

cross-sectional area of sample.

3.3.4. Abrasion Test

Abrasion tests were performed using an abrasion test machine (Atom Teknik AŞ.)

in accord with TS 669 standard [48]. Three samples of bricks of each composition

were involved for abrasion tests. Initial weight and height of the samples were

recorded prior to the test. The test procedure basically involves erosion of bottom

surface of test sample on a rotating disc by means of alumina powder laid the disc.

Nominal dimensions of abrasion test samples were 7 x 7 x 7 cm. The sample

remains stationary on the disc with 20 kgf axial pressure. The abrasion of the

sample occurred with the friction movement of the alumina powder between the

20

disc and the sample in a circular direction. A total of 22 rotations are set standard by

TS 699 for a single rotation direction. After 22 rotations, the disc was stopped and

test sample was removed. Disc surface was cleaned from previous dust residues.

A 20 g of alumina powder was laid for every 22 rotation step. Side direction of

sample was turned with 90º after each step. The test consists of 4 steps so that the

erosion occurring at the bottom of the sample would be uniform in all directions.

Differences of weight and height of the samples before and after abrasion test (after

the whole 4 steps) were recorded to obtain the values for abrasion resistance.

3.3.5. Water Resistance

Water durability tests were carried out on ten individual samples of each paste

composition. Samples were weighed and immersed in 100 ml of de-ionized water

for seven days. Weight of samples was measured using a Mettler Toledo brand

analytical balance (220 g capacity and 1 mg precision). After seven days duration,

samples were taken out from water and freshly weighed. The difference between

initial weight and weight after immersion was used as water absorption measure of

the pastes. The following formula was used to calculate water absorption rate.

After weighing, samples were dried in an oven at 65 ºC for two days to ensure all

absorbed water was evaporated. After two days of drying period samples were

weighed again. The difference between initial weight and dried weight was taken as

the measure of the weight loss in de-ionized water.

21

3.3.6. Polishing Tests

3.3.6.1. Polishing Test Samples

For polishing tests, coarse ground ceramic tiles supplied from Kaleseramik AŞ.

were used. Ceramic tiles were prepared in polishing facility of Kaleseramik AŞ. by

grinding the upper surface with 46 grit size abrasive bricks as calibration. Also, raw

ceramic tiles, which are only surface leveled with diamond disc, supplied from

Kaleseramik AŞ. were used for coarse grinding tests. Nominal dimensions of

ceramic tiles in length x width x depth were 30 x 30 x 1 cm, respectively.

Laboratory scale abrasive bricks for polishing performance tests were in cylindrical

shape. The cross-section area of the cylinder was 8 cm2 and height was 4 cm.

3.3.6.2. Experimental Setup for Polishing Tests

A laboratory scale polishing test setup was constructed to determine the polishing

performance of the bricks. Fundamental function of the setup was to simulate

industrial polishing process of ceramic tiles. The ceramic tile with nominal

dimensions of 30 x 30 x 1 cm in length x width x depth, respectively, was fixed to

bottom holder of the test machine. Above the ceramic tile, there were three satellites

holding three cylindrical abrasive bricks each. A simple schematic drawing of the

polishing test equipment is presented in Figure 3.1.

22

Figure 3.1. Schematic drawing of polishing test equipment.

The satellites rotated on the tile and around themselves simultaneously. By this

way, three groups of three abrasives had two simultaneous motions on the tile. Tap

water was pumped from the top of the machine both to cool the system and carry

out the polishing sludge. Under the bottom holder of the machine, pressure was

applied to the system by means of a pneumatic pump. Applied pressure and rotation

speed were adjusted and set with a control panel. After the test, a circular polished

path of about 5 cm wide and 15 cm outer diameter was obtained on the tile.

3.3.6.3. Polishing Test Parameters

The polishing tests were performed with the application of 2.5 bar pneumatic

pressure. Rotation speed of the motor was kept constant at 500 rpm. Polishing test

duration was 5 min.

23

3.3.6.4. Testing Procedure

Grinding (abrasion) and polishing tests were conducted in two different manners.

For grinding experiments, 46 grit size SiO2 powder embedded abrasive bricks were

applied on raw ceramic tiles that were leveled by diamond disc in Kaleseramik AŞ.

Surface roughness of tiles was measured using a profilometer. Initial roughness

value of raw ceramic tiles was not measured because the initial surface roughness of

the raw tiles was out of the measurement range of profilometer. After 46 grit size

grinding, 180 grit size SiO2 powder embedded abrasive brick was applied on the

same tile. Upon completion of the grinding test, surface roughness of tiles was

measured again and the results were recorded.

For polishing experiments, 46 grit size calibrated ceramic tiles supplied from

Kaleseramik AŞ. were used. Surface roughness of the ceramic tile was measured

before and after the polishing test. Then, 600 grade polishing test was carried out on

the same ceramic tile using 600 grit SiC powder added composite abrasive brick.

The data were recorded.

After the tests with 600 grade bricks, surface roughness of all ceramic tiles was

brought to almost the same roughness value to get a comparable data for the second

step polishing tests. In order to achieve equivalent degree of roughness, the ceramic

tiles were polished with the application of 25 wt% 600 grit SiC powder added

bricks for additional 5 min. Then, 1200 grade polishing tests were conducted. The

data obtained after each measurement were recorded. After the second step

polishing tests, ceramic tile surfaces were prepared for final polishing (1600 grade)

step. Surface roughness of all ceramic tiles was brought to almost the same

roughness value to get a comparable data for the final step polishing tests. In order

to achieve equivalent degree of roughness, the ceramic tiles were polished with the

application of 25 wt% 1200 grit SiC powder added bricks for additional 5 min.

Then, 1600 grade polishing tests were conducted by the use of neat MOC bricks.

The data obtained after each measurement were recorded.

24

3.3.6.5. Polishing Performance

3.3.6.5.1. Abrasive Consumption

The cylindrical neat MOC and composites samples (group of 9 samples for each

kind of composite brick) were weighed before and after each individual polishing

test. Difference between initial and final weight was considered as “weight loss”

occurred during the test. The weight loss of the bricks was evaluated as “Abrasive

Consumption” occurred during the 5 min of polishing test.

3.3.6.5.2. Surface Roughness

Surface roughness of the ceramic tiles was measured using a stylus profilometer

(Surtronic 3+, Taylor Hobson, England) before and after polishing tests.

A “TalyProfile Lite version: 3.1.4” (Taylor Hobson, England) software program

was employed to trace surface profile of the polished ceramic tiles. Ten

measurements were done on various locations of the polished tile to obtain reliable

data. The average of ten values was taken to establish mean roughness data.

Flowchart showing the experimental procedure of abrasive brick preparation is

presented in Figure 3.2.

25

Figure 3.2. Flowchart showing the experimental procedure of abrasive brick

preparation and testing.

26

CHAPTER 4

RESULTS AND DISCUSSION

In this section, the data gathered from the experimental study conducted on the

production and characterization of magnesium oxychloride cement (MOC) based

polishing bricks developed for polishing granite based ceramic tiles are presented.

The MOC based abrasive grinding and polishing bricks in various compositions

were formed, characterized, and tested to optimize the polishing performance of

ceramic tiles. Characterization, mechanical test, durability measurement, and

polishing test were performed according to the procedure as described in Section

3.3. Results obtained in this experimental study were presented, discussed and

assessed with the results of the previous studies conducted on the same subject and

published in the literature.

4.1. CHARACTERIZATION

Particle size distribution of 46 and 180 grit size SiO2 powders used in the present

study are illustrated in Figures 4.1 and 4.2, respectively.

27

Figure 4.1. Particle size distribution of 46 grit SiO2 powder.

Figure 4.2. Particle size distribution of 180 grit SiO2 powder.

Particle size distribution of 600 and 1200 grit size SiC powders are shown in

Figures 4.3 and 4.4, respectively.

Figure 4.3. Particle size distribution of 600 grit SiC powder.

28

Figure 4.4. Particle size distribution of 1200 grit SiC powder.

Particle size distribution analysis of MgO powder is shown in Figure 4.5.

Figure 4.5. Particle size distribution of MgO powder.

Results of particle size distribution analyses of SiO2, SiC and MgO powders are

summarized in Table 4.1.

Table 4.1. Particle size distribution data of raw materials.

Grit size d90 (µm) D50 (µm) D10 (µm) Surface area (m2/g)

46 (SiO2) 456 308 179 0.034

180 (SiO2) 276 195 136 0.064

600 (SiC) 20 12 7 0.551

1200 (SiC) 11 5 2 1.94

MgO 66 29 3 0.895

29

In addition to particle size distribution, surface area of the powders was measured

by laser diffraction technique. Results are presented also in Table 4.1. The analyses

revealed that 600 grit SiC powder had approximately four times larger surface area

than 1200 grit SiC powder. The data is quite relevant with geometrical principles.

That is to say, surface area is a two dimensional measure and particle size is a single

dimensional (i.e., diameter) measure. Thus, surface of an object changes with

square of the change occurring in particle size of the object. Surface area of MgO

powder was 0.895 m2/g, which is almost twice the surface area of 600 grit SiC

powder. Large surface area of active MgO powders is reported as a result of porous

structure in previous studies. Surface area and porosity are cited to be important

parameters for which active MgO is dissolved in MgCl2 solution but dead burnt

magnesia is not [22, 23].

X-Ray diffraction (XRD) pattern of MgO powder analyzed between 2θ of 5 to 75°

is given in Figure 4.6. The XRD peaks of crystalline MgO (JSPD # 45-0946) are

clearly seen. Small XRD peaks for SiO2 (JSPD # 46-1045) and MgCO3 (JSPD # 86-

2348) appear in the pattern. Chemical composition of MgO powder included 1.5

wt% SiO2, which is relevant to the results of XRD analysis. A small (~ 1 wt%)

amount of residual MgCO3 is due to low calcination temperature of caustic

magnesia [23, 30]. The XRD data obtained in this study are in accord with the

results reported in the literature.

Figure 4.6. X-Ray diffraction pattern of MgO powder.

30

XRD patterns of the MOC pastes 6-00, 7-00, and 8-00 are given in Figure 4.7. The

MOC F5 phase is indicated as F5 in the patterns. The XRD analyses of the MOC

pastes revealed that all three samples essentially composed of crystalline MOC F5

phase (JSPD # 07-0420) and some amount of residual MgO. XRD patterns in

Figure 4.7 exhibits two major peaks for MgO phase at 2θ of ~ 43° and ~ 62° which

is the major peak of MgO. Given that the XRD pattern of MgO powder exhibited all

three major peaks of MgO, the two MgO peaks in the MOC pastes is referred to as

residual MgO phase in cement pastes. Formation of MOC F5 phase was favored

but, formation of MOC F3 phase was not detected since the MgO/MgCl2 molar

ratio was more than 5. It has been reported that high concentrations of MgO

(MgO/MgCl2 molar ratio greater than 5) in the cement favors formation of F5 phase

and low MgO concentrations leads to the formation of F3 phase [16-19, 28-30].

Figure 4.7. X-Ray diffraction patterns of neat MOC pastes (from bottom to top

6-00, 7-00 and 8-00). Numbers on the curves indicate the MgO:MgCl2:H2O

molar ratio.

Findings of the XRD analyses are compatible with those reported in the literature

for the studies conducted on similar subjects. Li et al. [16] analyzed MOC pastes of

31

various molar compositions in which MgO/H2O ratio was changed for 1 mol

MgCl2. Figure 4.8 illustrates XRD patterns obtained in their study. When an

assessment is made between Figures 4.7 and 4.8, it is seen that the findings of this

study are in good agreement with the results presented by Li et al. [16].

Figure 4.8. XRD patterns of MOC pastes (from bottom to top, MgO/H2O ratio is

13/12, 13/16, 17/16, and 19/22).

Various researchers made use of intensity of XRD patterns for comparison of

crystallization behavior of cementious and ceramic materials [16, 27, 34, 49]. When

a comparison is made between the values of intensity of the peaks belonging to F5

phase in the MOC pastes, the intensity of F5 phase increased from paste 6-00 to

paste 8-00, implying that the formation of F5 phase increased with increasing

32

MgO/MgCl2 molar ratio in the paste. Moreover, the peaks of F5 phase sharpened

with increasing MgO/MgCl2 molar ratio in the paste as seen in Figure 4.7.

For the case of residual MgO phase, the comparison of XRD patterns reveals a

similarity between the peak intensity values of the pastes 6-00 and 7-00. However,

the XRD pattern of the paste 8-00 exhibits superior intensity peak for residual MgO

as compared to the patterns for the pastes 6-00 and 7-00. The results mean that a

higher amount of residual MgO phase is present in the paste 8-00.

4.2. DENSITY MEASUREMENTS

The change in density of the neat MOC pastes with regard to MgO/MgCl2 molar

ratio in the paste is shown in Figure 4.9. Error bars in the figure represent the ± 1

standard deviations from the average values of the determinations. It is evident that

density increases continuously with increasing MgO/MgCl2 molar ratio in the paste.

Figure 4.9. Variation in density of the neat MOC pastes with MgO/MgCl2 molar

ratio of the paste. The line is drawn to guide the eyes.

33

The density of the paste 6-00 was 1.90 g/cm3, but increased to 2.05 when

MgO/MgCl2 molar ratio in the MOC paste is increased to 8. The density of the

paste 7-00 which was used as binder phase in the production of abrasive bricks was

1.99 g/cm3. The increase in density of MOC pastes with increasing MgO/MgCl2

molar ratio implies that density increases also with increasing H2O/MgCl2 ratio

which was kept constant at 11/1 in this study. Cement/water ratio is considered as

an effective parameter on porosity and density of cement paste. Various

investigators reported that density increases as cement/water ratio is increased

[50-54]. Results of this study are accompanied with those reported in the literature.

The 46 and 180 grit size SiO2 powders were added in the amount of 30 wt% to the

paste 7-00 in order to prepare coarse grinding abrasive bricks. The densities of the

different mixtures were compared and shown in Figure 4.10. It is obvious that

density of the neat MOC paste increased with SiO2 powder addition. However, no

significant alteration in density of the 46 and 180 grit size SiO2 powder embedded

bricks was noticed. The density of both bricks was 2.09 g/cm3.

Figure 4.10. Density of different mixtures investigated in this study.

34

Densities of various amounts of SiC powder embedded abrasive bricks were

compared with the density of the neat MOC paste. The variation in density of the

600 and 1200 grit size SiC powder embedded abrasive bricks with SiC addition is

shown in Figure 4.11. Error bars in the figure represent the ± 1 standard deviations

from the average values of the determinations. It is obvious that density of the neat

MOC paste increased also with SiC powder additions. The density of the neat MOC

paste increased from 1.99 to 2.04 g/cm3 when 20 %wt 600 grit size SiC powder is

embedded. The densities of the 25 and 30 %wt 600 grit size SiC powder embedded

polishing bricks were 2.05 and 2.06 g/cm3, respectively.

The densities of the 1200 grit size SiC powder embedded polishing bricks were not

much different than those of the 600 grit size SiC powder embedded polishing

bricks. The densities of the 20, 25 and 30 %wt 1200 grit size SiC powder embedded

polishing bricks were 2.04, 2.05, and 2.05 g/cm3, respectively.

Figure 4.11. Variation in density of 600 and 1200 grit size SiC powder embedded

abrasive bricks with SiC powder addition. The lines are drawn to guide the eyes.

35

Density measurements revealed that there is a limiting value for density of the SiC

powder embedded polishing bricks. That is to say, when the SiC powder addition is

increased beyond 25 wt%, further increase in density was not attained in 1200 grit

size SiC powder embedded abrasive bricks. The change in 600 grit size SiC powder

embedded abrasive bricks was within the error limits. This phenomenon is probably

related to viscosity and porosity of cement paste. As stated earlier in Section 3.3.4,

high amount of fine SiC powder additions to cement pastes decreases workability

and increase viscosity [53]. High viscosity probably causes air bubbles to be

entrapped in the paste but not reach the free surface. Consequently, additional SiC

powder increases viscosity which in turn hinders the increase in density above a

certain limit.

A similar phenomenon of viscosity is observed for a cementious material by Zhang

and Han who stated that ultra-fine admixtures such as silica fume and power plant

fly ash are effective on rheological properties of cement paste [55]. A low amount

of ultra fine powder addition to cement paste was found to improve the rheological

properties of cement paste in terms of viscosity and yield strength.

4.3. COMPRESSION TESTS

The variation in compressive strength of the neat MOC pastes with increasing

MgO/MgCl2 molar ratio in the paste is illustrated in Figure 4.12. Error bars in the

figure represent the ± 1 standard deviations from the average values of the

determinations. Compressive strength of the neat MOC pastes increased with

increasing MgO/MgCl2 molar ratio. The values for the compressive strength of the

pastes 6-00, 7-00, and 8-00 were 53.8, 74.2, and 86.5 MPa, respectively.

36

Figure 4.12. Variation in compressive strength of the MOC pastes with MgO/MgCl2

molar ratio in the paste. The line is drawn to guide the eyes.

Compressive strength of MOC is widely investigated by several researchers [16, 18,

19, 23, 30, 42] who clarified two main parameters; MgO/MgCl2 molar ratio and

H2O/MgCl2 ratio, to be effective on strength of MOC. As MgO/MgCl2 molar ratio

increases, compressive strength of MOC increases. Harper [23] studied a wide

range of molar compositions with three different MgO powders and found similar

relation between MgO/MgCl2 ratio and compressive strength. He reported 60 to 80

MPa compressive strength range for the MOC having an MgO/MgCl2 ratio of 7/1.

Li et al. [16] also studied compositional dependence of compressive strength of

MOC. They reported compressive strength of a MOC paste with MgO/MgCl2 molar

ratio of 7/1 and H2O/MgCl2 ratio of 10/1 as 110 MPa. Özer et al. [56] presented 28

day-compressive strength of a MOC with MgO/MgCl2 molar ratio of 7/1 as 28.8

MPa. The deviation in compressive strength of MOCs among the reported values in

the literature is due to the dissimilarity in raw materials used for the preparation of

MOC pastes. Indeed, various investigators stated the importance of especially MgO

properties on the strength of MOC pastes [19, 23]. Calcination temperature and

particle size of MgO powder were found to be among the most important factors for

setting and hardening of MOC pastes.

37

The values of the compressive strength of SiO2 or SiC powder embedded abrasive

bricks were greater than those of the MOC pastes. Figure 4.13 presents the

compressive strength values of 30 wt% 46 and 180 grit size SiO2 powder embedded

abrasive grinding bricks. Compressive strength of paste 7-00 increased from 74.2

MPa to 103.5 and 105.2 MPa with 30 wt% 46 and 180 grit SiO2 powder addition,

respectively. However, values of the compressive strength of the 46 and 180 grit

size SiO2 powder embedded bricks, were not significantly different from each other.

Coarse and fine aggregate additions to MOC pastes were extensively studied by

several researchers who recognized that additions of coarse and fine aggregates

enhance mechanical and physical properties of MOC [54-55, 57-61].

Figure 4.13. Compressive strength of different mixtures investigated in this study.

Compressive strengths of various amounts of SiC powder embedded abrasive bricks

were compared with that of the paste 7-00. The variation in compressive strength of

the 600 and 1200 grit size SiC powder embedded abrasive bricks with SiC powder

38

addition is shown in Figure 4.14. It is apparent that compressive strength of the neat

MOC paste increases with increasing SiC powder additions.

Figure 4.14. Variations in compressive strength of the 600 and 1200 grit size SiC

powder embedded abrasive bricks with SiC powder addition. The lines are drawn to

guide the eyes.

The compressive strength of neat MOC paste 7-00 increased from 74.2 MPa to 92.4

MPa when 20 %wt 600 grit SiC powder is embedded. The compressive strengths of

the 25 and 30 %wt 600 grit SiC powder embedded polishing bricks were 98.5 MPa

and 97.2 MPa, respectively. The compressive strength of the 1200 grit SiC powder

embedded polishing bricks exhibited similar tendency to the 600 grit SiC powder

embedded polishing bricks. Compressive strengths of the 20, 25 and 30 %wt 1200

grit SiC powder embedded polishing bricks were 89.4, 95.6, and 94.0 MPa,

respectively.

39

Experimental results reveal that compressive strength of the MOC paste increases

with both 600 and 1200 grit size SiC powder additions. The difference in the

increase in compressive strength between 600 grit and 1200 grit SiC powder

embedded MOCs is tied to the differences in particle size and surface area of these

powder. Particle size of 600 grit SiC powder is twice the particle size of 1200 grit

SiC powder. Specific surface area of 600 grit SiC powder is four times larger than

that of 1200 grit SiC powder as seen in Table 4.1 in Section 4.1. Particle size and

specific surface area have influence on hydration and setting reactions of cements

[62]. Very fine particles of SiC powder trap more capillary water and yields

microporosities in cement structure. Coarser particles of SiC powder on the other

hand, contributes soundness of three dimensional MOC network by filling voids

and increase compressive strength [61-63]

As seen in Figure 4.14, compressive strength values exhibited retardation over 25

wt% addition of SiC powders. Specifically, when SiC powder content is increased

beyond 25 wt%, additional increase in compressive strength was not acquired due

to binder/aggregate ratio is decreased. The effect of fine SiC powders on

compressive strength of MOC pastes is in good agreement with the results of

previous works [54, 58-60].

4.4. ABRASION TESTS

Abrasion resistance of the neat MOC pastes and SiO2 or SiC embedded abrasive

bricks was measured according to dry abrasion test as described in Section 3.3.4.

Weight loss and decrease of height due to abrasion were determined and noted as

test data. Figure 4.15 shows the variation of weight loss and height loss occurred

during abrasion test of the neat MOC pastes as a function of MgO/MgCl2 molar

ratio in the paste. Error bars in the figure represent the ± 1 standard deviations from

the average values of the determinations. As seen in Figure 4.15, weight loss and

height loss occurred due to abrasion decreased with increasing MgO/MgCl2 molar

40

ratio up to 7/1. Beyond that weight loss increased but height loss did not change.

Weight loss occurred in the pastes 6-00, 7-00, and 8-00 was 48.6, 31.6, and 33.8 g,

respectively. Height loss of paste 6-00 was 4.9 mm. Height loss occurred in pastes

7-00 and 8-00 was about the same, 3.1 mm.

Figure 4.15. Variation in weight loss and height loss occurred during abrasion test

of the neat MOC pastes with MgO/MgCl2 molar ratio in the paste. The lines are

drawn to guide the eyes.

Abrasion tests have revealed that abrasion resistance of the neat MOC pastes exhibit

also an inflection point at MgO/MgCl2 molar ratio of 7/1. When a comparison is

made between the height loss values of neat MOC pastes, the paste 6-00 undergoes

highest abrasion than the pastes 7-00 and 8-00. Paste 8-00 was prone to slightly

higher weight loss compared to paste 7-00, although both had same height loss

during abrasion tests. Indeed, the height loss of pastes 7-00 and 8-00 were small

numbers and weight loss values were very close, which makes the comparison

difficult. The reason why the paste 8-00 has higher abrasion than the pastes 7-00

can be attributed to the difference in density of the pastes. Since density of the paste

8-00 is greater than the paste 7-00, weight loss upon abrasion of the paste 8-00 was

41

measured to be greater than paste 7-00. Consequently, the paste 7-00 showed best

abrasion resistance.

Abrasion resistance of SiO2 or SiC powder embedded abrasive bricks was better

than that of the MOC pastes. Figure 4.16 presents the weight loss and height loss

values of 30 wt% 46 and 180 grit size SiO2 powder embedded abrasive bricks.

Weight loss value of paste 7-00 decreased from 31.6 g to 12.4 and 13.0 g with 30

wt% 46 and 180 grit SiO2 powder addition, respectively. Height value of paste 7-00

also decreased from 3.1 mm to 1.1 and 1.2 mm with 30 wt% 46 and 180 grit SiO2

powder addition, respectively. Addition of SiO2 powders of both size to the MOC

pastes improved abrasion resistance of MOC considerably. However, weight loss

and height loss values of SiO2 embedded abrasive grinding bricks were very close

to each other. As mentioned earlier in Section 4.3, additions of coarse and fine

aggregate to MOC pastes enhance mechanical and physical properties of MOC

[54, 55, 57-61].

Figure 4.16. Variation in weight loss and height loss occurred during abrasion tests

with SiO2 powder addition to neat MOC. The lines are drawn to guide the eyes.

Abrasion resistances of various amounts of SiC powder embedded abrasive bricks

were compared with that of the paste 7-00. The change in abrasion resistance in

42

terms of weight and height loss of the 600 and 1200 grit size SiC powder embedded

abrasive bricks with SiC addition are given in Figures 4.17 and 4.18, respectively. It

is noticeable that abrasion resistance of the neat MOC paste increases with

increasing SiC powder additions.

Figure 4.17. Variation of weight loss and height loss occurred during abrasion tests

of the 600 grit size SiC powder embedded abrasive bricks with SiC powder

addition. The lines are drawn to guide the eyes.

Figure 4.18. Variation of weight loss and height loss occurred during abrasion tests

of the 1200 grit size SiC powder embedded abrasive bricks with SiC powder

addition. The lines are drawn to guide the eyes.

43

Weight loss and height loss due to abrasion of paste 7-00 was 31.6 g and 3.1 mm,

respectively. Weight loss of neat MOC paste 7-00 decreased from 31.6 g to 16.8 g

when 20 %wt 600 grit SiC powder is admixed. Weight loss of the 25 and 30 %wt

600 grit SiC powder embedded polishing bricks was 12.1 and 13.4 g, respectively.

Height loss of the 600 grit SiC powder added abrasive bricks were 1.6, 1.1 and 1.3

mm for the 20, 25 and 30 wt% SiC powder additions, respectively. The 1200 grit

SiC powder embedded polishing bricks displayed similar abrasion behavior with the

600 grit SiC powder embedded polishing bricks. In the case of 1200 grit SiC

powder addition to paste 7-00, weight loss values were 20.7, 18.5 and 19.3 g for the

20, 25 and 30 wt% SiC powder additions, respectively. Height loss values of the

1200 grit SiC powder embedded abrasive bricks were 2.8, 1.8 and 1.9 mm for the

20, 25 and 30 wt% SiC powder additions, respectively.

The experimental findings presented in Figures 4.15 through 4.18 reveal that

abrasion resistance of MOC based abrasive bricks are profoundly affected with the

three parameters: 1) MgO/MgCl2 molar ratio of binder phase, 2) particle size and

type of abrasive powder addition, and 3) amount of SiC powder addition. Abrasion

resistance of the neat MOC pastes increased with increasing of MgO/MgCl2 molar

ratio, which was found to be a determinant factor on compressive strength of MOC

in Section 4.3.

Unfortunately, there are only a few studies [18, 63-65] and very limited data on

abrasion resistance of the MOC cements from which the results of this study could

be assessed directly. However, Bauzoubaa et al. [63] reported 1.3 to 1.6 mm range

of abrasion depth (which is defined as height loss in this publication) for fine

aggregate embedded cements. Height loss values of the 600 grit SiC powder added

abrasive bricks were within this range but that of the 1200 grit SiC powder added

abrasive bricks were somehow superior. Addition of either SiO2 or SiC powder to

the MOC pastes was found to increase abrasion resistance. Figures 4.17 and 4.18

imply an inflection point for SiC powder additions. Small additions of SiC powder,

such as 20 and 25 wt%, improved abrasion resistance. When SiC powder additions

44

were made more than 25 wt%, abrasion resistance was weaken. The decrease in the

abrasion resistance with increasing amount of SiC powder is probably due to

decrease in the amount of binder phase, which result in weakening of the structure.

Experimental findings of abrasion tests are well-matched with those of previous

works [55, 63].

4.5. WATER RESISTANCE

Water resistance of the neat MOC pastes and SiO2 or SiC powder embedded

abrasive bricks was calculated from the values obtained through measurements of

water absorption and weight loss after immersion in de-ionized water for 7 days.

Figure 4.19 shows the variation in water absorption and weight loss occurred during

immersion test of the neat MOC pastes as a function of MgO/MgCl2 molar ratio in

the paste. Error bars in the figure represent the ± 1 standard deviations from the

average values of the determinations.

Figure 4.19. Variation in water absorption and weight loss after immersion of the

neat MOC pastes with MgO/MgCl2 molar ratio in the pastes. The lines are drawn to

guide the eyes.

45

As seen in Figure 4.19, water absorption and weight loss occurred in the neat MOC

pastes decreased with increasing MgO/MgCl2 molar ratio up to 7/1. Beyond that

water absorption and weight loss seemed unaffected with MgO/MgCl2 molar ratio.

Water absorption of the neat MOC pastes 6-00, 7-00 and 8-00 was 1.81, 1.28 and

1.17 wt%, respectively. Weight loss of the pastes in the previous order was 5.76,

2.97 and 2.83 wt%, respectively. Both water absorption and weight loss of neat

MOC pastes upon immersion into de-ionized water decreased with increasing

MgO/MgCl2 molar ratio in the pastes. This is attributed to the structure developed

in the MOC pastes. As the amount of MgO in the MOC paste is increased, the

structure becomes more compact and less permeable to aqueous media. Also,

crystallization of F5 phase, which is known with high strength and durability among

the other MOC phases, is promoted with high amounts of MgO. Lastly, excess

MgO in the structure of MOC serves as a blanket for protecting the F5 crystals from

aqueous media [19, 29, 34]. Consequently, water resistance of MOC increases with

increasing of MgO/MgCl2 molar ratio.

In addition to neat MOC pastes, SiO2 or SiC powder embedded abrasive bricks

were subjected to water resistance test. Water absorption and weight loss of SiO2

embedded abrasive bricks were calculated. Results of the water resistance test are

shown graphically in Figure 4.20.

Figure 4.20. Water absorption and weight loss of different mixtures investigated in

the present study. The lines are drawn to guide the eyes.

46

Error bars in the figure represent the ± 1 standard deviations from the average

values of the determinations. Water absorption and weight loss occurred during

immersion tests of the 46 grit size SiO2 powder embedded abrasive bricks were 1.04

and 1.74 wt%, respectively. In the case for the 180 grit SiO2 powder addition, the

values were determined as 0.95 and 1.46 wt%, respectively.

Water resistance of various amounts of SiC powder embedded abrasive bricks was

compared with that of the paste 7-00. The change in water resistance in terms of

water absorption and weight loss in de-ionized water for the 600 and 1200 grit size

SiC powder embedded abrasive bricks with SiC addition are given in Figures 4.21

and 4.22, respectively. Error bars in the figure represent the ± 1 standard deviations

from the average values of the determinations. It is clearly seen that water resistance

of the neat MOC paste increases with increasing SiC powder additions.


immersion tests of the 600 grit size SiC powder embedded abrasive bricks with SiC

powder addition. The lines are drawn to guide the eyes.

Water absorption and weight loss due to immersion of paste 7-00 was 1.28 and 2.97

wt%, respectively. Water absorption of neat MOC paste 7-00 decreased from 1.28

wt% to 0.63 wt% when 20 wt% 600 grit size SiC powder is added. Water

absorption of the 25 and 30 %wt 600 grit size SiC powder embedded polishing

47

bricks was 0.42 and 0.18 wt%, respectively. Weight loss values of the 600 grit size

SiC powder added abrasive bricks were 2.27, 2.02 and 2.32 wt% for the 20, 25 and

30 wt% additions, respectively.


immersion tests of the 1200 grit size SiC powder embedded abrasive bricks with

SiC powder addition. The lines are drawn to guide the eyes.

The 1200 grit SiC powder embedded polishing bricks displayed similar water

absorption behavior with the 600 grit SiC powder embedded polishing bricks. In the

case of 1200 grit SiC powder addition to paste 7-00, water absorption values were

0.65, 0.38, and 0.14 wt% for the 20, 25 and 30 wt% SiC powder additions,

respectively. Weight loss values of the 1200 grit SiC powder embedded abrasive

bricks were 2.59, 2.21 and 2.45 wt% for the 20, 25 and 30 wt% SiC powder

additions, respectively.

Water resistance measurements revealed that both 46 and 180 grit SiO2 powder

addition in the amount of 30 wt% improved water resistance of the MOC paste.

Figures 4.21 and 4.22 illustrate an inflection point for SiC powder addition at 25

wt% where weight loss of abrasive brick increases although water absorption value

decreases. The reason of inflection point is explained with the formula to calculate

48

the water absorption rate given in Section 3.3.5. In practice, there is an amount of

mass loss which is hidden in the “weight after immersion” term due to Cl- ion

release into de-ionized water during immersion duration [17, 19, 34, 35]. Thus, the

water absorption rate is not enough an accurate data on itself, however, it can be

used to estimate permeability characteristics of cements [51, 58-61, 66].

4.6. POLISHING TESTS

Polishing performance of neat MOC pastes and SiO2 or SiC powder embedded

abrasive bricks was measured as described in section 3.3.6. Surface roughness of

ceramic tiles and abrasive brick consumption were taken as performance

parameters.

First, grinding tests were performed with the application of abrasive brick 7-Q-46

(46 grit SiO2 powder embedded brick) on raw ceramic tiles. Upon completing the

grinding process with 7-Q-46 abrasive brick, surface roughness of the tile was

measured and noted. Then, abrasive brick 7-Q-180 was applied on the same tile.

Upon completing the grinding process with 7-Q-180 abrasive brick, surface

roughness of the tile was measured again. The data were noted and used to evaluate

the grinding performance of the abrasive bricks. Mean surface roughness of ceramic

tiles and abrasive erosion occurred during the application of 46 and 180 grit

grinding practices are presented in Figure 4.23.

Figure 4.24 illustrates representative surface profiles of ceramic tiles upon

completion of 46 and 180 grit grinding practices. Mean surface roughness of

ceramic tile was 2.25 µm after the application of abrasive brick 7-Q-46 for 5 min.

The consumption occurred in brick 7-Q-46 was 159.5 g. Mean surface roughness of

the ceramic tile decreased to 1.39 µm after the application of abrasive brick

7-Q-180. The consumption occurred in brick 7-Q-180 was 61.8 g. When a

comparison is made between the mean surface roughness and abrasive consumption

49

values of the abrasive bricks 7-Q-46 and 7-Q-180, erosion of the abrasive brick

7-Q-46 was almost three times greater than that of the brick 7-Q-180.

Figure 4.23. Mean surface roughness of ceramic tiles and abrasive consumption

occurred during the application of 7-Q-46, 7-Q-180, and 7-25-C coded abrasive

bricks. The lines are drawn to guide the eyes.

Figure 4.24. Surface roughness profile of ceramic tile after the application of

abrasive bricks, a) 7-Q-46 and b) 7-Q-180.

50

Upon completing the grinding procedures with abrasive bricks 7-Q-46 and 7-Q-180,

polishing tests were performed on 46 grit size calibrated ceramic tiles supplied from

Kaleseramik AŞ. Polishing processes began with the application of 600 grit size

SiC powder embedded abrasive bricks containing 20, 25 and 30 wt % SiC powder

(coded as 7-20-C, 7-25-C, and 7-30-C, respectively). Then, 1200 grit size SiC

powder embedded abrasive bricks containing same amounts of SiC powder (coded

as 7-20-F, 7-25-F and 7-30-F, respectively) were applied on the same tile. Polishing

processes ended with the application of the neat MOC paste 7-00 on the same tile.

Surface roughness of ceramic tiles was measured before and after application of

each bricks.

Data obtained from polishing tests were plotted in terms of surface roughness

against grit number of the SiC powder embedded in the brick for different amounts

of SiC powder additions. Mean surface roughness of ceramic tiles after being

exposed to different polishing practices with regard to abrasive grit size is shown in

Figure 4.25. The initial data point is for the mean roughness value of 46 grit

calibrated raw ceramic tile supplied from Kaleseramik AŞ.

Figure 4.25. Mean surface roughness of ceramic tiles after being exposed to

different polishing procedures with regard to grit size of the abrasive embedded.

The lines are drawn to guide the eyes.

51

Mean surface roughness of ceramic tiles decreased successively in different steps of

the polishing procedures. For 600 grit polishing practices, mean surface roughness

of the tile was measured as 0.95, 0.56, and 0.62 µm for the bricks 7-20-C, 7-25-C,

and 7-30-C, respectively. In the case of application of the 1200 grit size abrasive

bricks, mean surface roughness of the tile was 0.22, 0.27, and 0.28 µm for the bricks

7-20-F, 7-25-F, and 7-30-F, respectively. Figure 4.26 illustrates representative

surface profiles of ceramic tiles upon completion of 600 and 1200 grit polishing

practices.

Figure 4.26. Sample roughness profile of ceramic tile after polishing with a) 7-25-C

(mean roughness is 0.56 µm) and b) 7-25-F (mean roughness is 0.27 µm).

In addition to surface roughness of ceramic tiles, the consumption of the bricks

during polishing practices was determined to evaluate the polishing performance of

bricks. The consumption occurred in the 600 and 1200 grit size SiC powder

embedded bricks are shown in Figure 4.27. Abrasive consumption occurred in the

52

neat cement paste 7-00 during polishing is indicated also in Figure 4.27 for

comparison purpose.

Figure 4.27. Variations in abrasive brick consumption with grit size of SiC powder

for different amounts of SiC powder addition. The lines are drawn to guide the eyes.

The values of the abrasive consumption occurred in the 600 grit size SiC powder

embedded bricks were 89.5, 73.8, and 70.2 g for the bricks 7-20-C, 7-25-C, and

7-30-C, respectively. As for the 1200 grit size SiC powder embedded bricks, the

values of the abrasive consumption were 35.1, 27.5, and 26.2 g for the bricks 7-20-

F, 7-25-F, and 7-30-F, respectively. When a comparison is made between the values

of the abrasive consumption occurred in the bricks containing the same amount of

SiC powder, e.g. 7-20-C and 7-20-F, it is seen that erosion occurred in the 600 grit

size SiC powder embedded bricks was almost three times greater than that of the

1200 grit size SiC powder embedded bricks. In other words, erosion of abrasive

bricks decreased from coarse to fine SiC powder addition. Figure 4.23 represents a

similar trend for SiO2 embedded abrasive bricks. Actually, the abrasion resistance

of coarse powder embedded bricks was found to be better than fine powder

embedded bricks in Section 4.3. The difference in the erosion between different grit

size SiC powder embedded abrasive bricks is attributed to the initial surface

roughness of the ceramic tile not abrasion resistance of the bricks [3, 14].

53

The initial roughness of raw ceramic tile that was leveled with diamond disc is

reported to be more than 100 µm [11, 12, 14]. After the application of brick 7-Q-46,

the roughness value decreased drastically implying high amount of material

removal from ceramic surface [12-14]. The mean surface roughness value of the tile

before the application of brick 7-Q-180 was 2.25 µm which decreased to 1.39 µm

after the application of brick 7-Q-180. As a result, material removal and erosion of

brick 7-Q-180 was less than those of brick 7-Q-46. Similarly, erosion in 600 grade

polishing bricks was foremost among all the bricks tested, although the 600 grit size

SiC powder embedded bricks were found to have greatest abrasion resistance. The

reason why abrasive consumption was the foremost in 600 grade polishing practices

is tied to the fact that most of the roughness reduction (smoothening) of ceramic tile

surface is achieved in 600 grade polishing practice.

The aim of present study was to understand and evaluate the polishing performance

of abrasive bricks. Thus, abrasive brick types should be compared for each grit size

individually. For 600 grit size SiC embedded abrasive bricks, abrasive consumption

of 7-20-C was the greatest, 89.5 g. Bricks 7-25-C and 7-30-C exhibited more or less

the same amount of erosion (70.2 and 73.8 g, respectively) as seen in Figure 4.26.

Considering the surface properties of ceramic tiles after 600 grit polishing practices,

surface roughness was the greatest after exposing brick 7-20-C, 0.95 µm. Surface

roughness of the bricks 7-25-C and 7-30-C were close to each other having values

of 0.56 and 0.62 µm, respectively.

Polishing tests measurements revealed that among the abrasive bricks containing

identical SiC powder, e.g. coarse or fine, the bricks containing 25 wt% SiC powder

exhibited the best polishing performance in terms of both roughness reduction and

abrasive consumption. There is very limited data reported on polishing performance

of abrasive bricks with respect to abrasive composition in order to support and/or

compare this judgment. Findings of polishing practices achieved in this study in

terms of surface roughness and abrasive consumption values are consistent with

those reported in previous publications [3, 12-14].

54

Three different types of neat cement pastes namely; 6-00, 7-00, and 8-00 were

applied for the last step (1600 grade) polishing practices. Variation in the final

surface roughness of ceramic tiles and abrasive brick consumption after 1600 grade

polishing practices with MgO/MgCl2 molar ratio is illustrated in Figure 4.28.

Figure 4.28. Variation in surface roughness of ceramic tile and abrasive brick

consumption upon 1600 grade polishing practice with MgO/MgCl2 molar ratio in

the paste. The lines are drawn to guide the eyes.

Representative surface profiles of ceramic tile upon completion of 1600 grit

polishing practices with the neat MOC pastes 6-00, 7-00 and 8-00 are given in

Figure 4.29. Surface roughness of ceramic tile after being polished with paste 6-00

was 0.18 µm. It decreased to 0.12 µm after being polished with paste 7-00, but did

not change much after being polished with paste 8-00. Abrasive brick consumption

after the polishing practices was 25.5, 15.4, and 16.2 g for the pastes 6-00, 7-00, and

8-00, respectively.

Erosion of paste 6-00 due to polishing was the greatest among the neat MOC pastes.

The bricks made from paste 8-00 exhibited slightly more consumption than those

made from paste 7-00 although both bricks resulted in more or less the same

smoothness on the tile. The reason of high abrasive consumption in paste 8-00

55

compared to paste 7-00 is attributed to polishing test duration [3]. Surface

roughness and glossiness of the ceramic tiles on the other hand, is limited with

structural and compositional properties of tile itself [6, 10, 39]. Excess of polishing

durations especially for final grade (1600 grit) does not decrease surface roughness

but increases abrasive consumption [3, 14, 39].

Figure 4.29. Surface roughness profile of ceramic tile after being polished with

paste a) 6-00 (mean roughness is 0.18 µm), b) 7-00 (mean roughness is 0.12 µm)

and c) 8-00 (mean roughness is 0.12 µm).

56

4.7. MICROSTRUCTURAL EXAMINATIONS

Microstructure of the neat MOC pastes and of the surface of the ceramic tiles

polished with polishing disks were examined on freshly fractured or polished

surfaces by Scanning Electron Microscope (SEM) according to the procedure as

described in Section 3.3.1. Microstructural analyses were done to obtain

supplementary information in order to evaluate the chemical and mechanical

behavior observed during the tests.

4.7.1. Microstructure of the Neat MOC Pastes

A representative general view of the microstructure for the neat MOC pastes 6-00,

7-00, and 8-00 are shown in Figures 4.30 (a), (b), and (c), respectively.

(a)

Cont…

57

(b)

(c)

Figure 4.30. SEM fractographs of the neat MOC pastes of (a) 6-00, (b) 7-00,

(c) 8-00.

58

SEM investigations have revealed that paste 6-00 has a porous microstructure.

Micro porosities dispersed in the structure of paste 7-00 are apparent as seen in

Figure 4.30 (b). Nonetheless, paste 8-00 exhibited a dense and compact structure, as

indicated in 4.30 (c). The findings of the microstructure examinations are in good

agreement with the density measurement and water absorption test results presented

in Sections 4.2 and 4.3, correspondingly. Density is related to porosity [54, 56, 68].

As discussed earlier in Section 4.2, the density of the neat MOC pastes varies with

MgO/MgCl2 molar ratio. Also, the amount of water absorption of paste 6-00, which

possessed least MgO/MgCl2 molar ratio among the neat MOC pastes investigated,

was more than that of pastes 7-00 and 8-00.

Consistent with the results given in the literature [22, 19, 29], crystalline phase

development in neat MOC pastes was generally in needle shape. The needle shaped

crystals were visible mostly in micro porosities of the pastes 6-00, 7-00, and 8-00 as

shown in Figures 4.31 (a), (b), and (c), respectively.

(a)

Cont…

59

(b)

(c)

Figure 4.31. SEM micrographs of the needle shaped MOC crystals in pastes of

(a) 6-00, (b) 7-00, (c) 8-00.

60

Porous structure of paste 6-00 is apparent in Figure 4.31 (a). The neat MOC crystals

spread in loose form. Needle shaped MOC crystals are also visible in micrographs

of pastes 7-00 and 8-00 as shown in Figure 4.31 (b) and (c), respectively. The

needle shaped crystals were identified as MOC F5 phase in characterization studies

presented in Section 4.1.

4.7.2. Surface Microstructure of Polished Ceramic Tiles

Polished surface of ceramic tiles was examined using SEM after polishing practices

to determine the polishing performance of the abrasive bricks. Indeed, SEM

analysis itself is not accurate enough to resolve the polishing performance of

abrasive bricks, still could be used as supportive information to roughness and gloss

measurements [3, 6, 10]. Therefore, microstructural observations were taken as

supplementary information along with roughness measurements to understand and

estimate the polishing performance of abrasive bricks. 600 grade polishing practices

were applied on raw (ground with 46 grit abrasive) ceramic tiles. The abrasive

bricks containing various amounts of 600 grit size SiC powder, e.g. bricks of 7-20-

C, 7-25-C, and 7-30-C, were employed for this step of polishing practices. Figure

4.32 shows microstructures of ceramic tile surface after being exposed to 600 grit

polishing practices with abrasive bricks containing various amounts of 600 grit size

SiC powder.

SEM micrographs in Figure 4.32 exhibited the presence of open porosities on

ceramic tiles. Also, scratches and polishing traces are visible in the images.

61

(a)

(b)

Cont…

62

(c)

Figure 4.32. SEM micrographs of ceramic tile surface after being exposed to 600

grit polishing practice with abrasive bricks of a) 7-20-C, b) 7-25-C and c) 7-30-C.

1200 grade polishing practices were done with the application of 1200 grit size SiC

powder embedded abrasive bricks. SEM micrographs taken from polished surface

of ceramic tile after being exposed to abrasive bricks containing various amounts of

1200 grit size SiC powder are given in Figure 4.33. Micro porosities and polishing

traces are visible in the SEM images of ceramic tiles. When a comparison is made

between the images in Figure 4.32 with those in Figure 4.33, it is clear that tiles

exposed to 600 grit polishing practices show evidences of more and better

noticeable traces than those exposed to 1200 grit polishing practices.

63

(a)

(b)

Cont…

64

(c )

Figure 4.33. SEM micrographs taken from polished surface of ceramic tile after

being exposed to 1200 grit polishing practice with abrasive bricks of a) 7-20-F,

b) 7-25-F and c) 7-30-F.

SEM micrographs in Figure 4.34 show microstructures of ceramic tiles after final

polishing practices with the neat MOC pastes. Figure 4.34 (b) reveals that polishing

traces and scratches are removed from ceramic tile surface after polishing it with

paste 7-00. However open porosities on the surface of ceramic tiles, which are

intrinsic property of ceramic tiles, are evident. Esposito et al. [11] studied

microstructure of polished ceramic tiles and reported plastic deformation of ceramic

tile surface due to polishing process. They also mentioned opening of micro

porosities in the microstructure as a drawback of polishing operation.

Microstructures observed in the present work are well-suited with that observed by

Esposito et al [11].

65

(a)

(b)

Cont…

66

(c)

Figure 4.34 SEM micrographs taken from polished surface of ceramic tile after

being exposed to the neat MOC of a) paste 6-00, b) paste 7-00 and c) paste 8-00.

Combining the results obtained from roughness measurements and SEM

examinations of polished ceramic tiles, it is clear that the neat MOC paste 7-00

exhibited best qualifications in terms of polishing performance.

67

CHAPTER 5

SUMMARY and CONCLUSIONS

Magnesium oxychloride cement (MOC) based grinding and polishing bricks

developed for polishing of granite based ceramic tiles were produced and

characterized. 46 or 180 grit size SiO2 powder embedded MOC based abrasive

bricks were applied on granite ceramic tiles to accomplish surface grinding. For

polishing practices 600 and 1200 grit size SiC powder embedded MOC based

polishing bricks and neat (unreinforced) MOC pastes were exposed on granite

ceramic tiles.

The bricks were prepared by conventional technique of mixing and molding of the

cement paste. Three different neat MOC pastes depending on MgO/MgCl2 molar

ratio in the paste were formed. Grinding bricks were formed by adding 30 weight

percentage, wt%, of both of the SiO2 powders. Polishing bricks were formed by

adding 20, 25, and 30 wt% of each SiC powders. Crystalline phase formation and

microstructure of the neat cements were examined using X-Ray powder diffraction

and scanning electron microscope, (SEM). Mechanical properties namely;

compressive strength and abrasion resistance, chemical durability in water and

polishing ability of the neat MOC pastes and abrasive bricks were determined to

improve the polishing performance. Density, water resistance, and abrasion

resistance of MOC paste increased with increasing MgO/MgCl2 molar ratio in the

paste. Improvement in water resistance was due to promotion of the formation of F5

crystals and low porous compact structure in MOC pastes.

68

Additions of either SiO2 or SiC powders had a profound effect on compressive

strength and water resistance of MOC pastes. Increasing additions of fine and

coarse SiC powders up to 25 wt% enhanced mechanical properties. More than 25

wt% addition of SiC powders had a tendency to decrease the compressive strength

and water resistance of MOC paste.

The MOC paste composed of 7:1:11 molar ratio of MgO:MgCl2:H2O was found to

have greatest resistance to abrasion. Type and amount of abrasive powder addition

had a powerful effect on abrasion resistance. Additions of either SiO2 or SiC

powders improved abrasion resistance of MOC paste, too. Advancement in the

abrasion resistance of MOC with 30 wt% 46 or 180 grit size SiO2 powders addition

was more or less the same. Additions of 600 or 1200 grit size SiC powders up to a

limiting value of 25 wt% led to an improving effect on abrasion resistance. For a

given amount of addition, coarse SiC powders were more effective than fine SiC

powders.

Polishing performance of abrasive bricks was evaluated in terms of mean surface

roughness and abrasive brick consumption occurred during polishing. SEM

examinations revealed the evidences of the reasons that 25 wt% SiC powder

embedded abrasive bricks has the best qualifications in terms of abrasion resistance

and polishing performance.

69

REFERENCES

1. C. P. M. Tenorio, M. Dondi, G. Ercolani, G. Guarini, C. Melandri, M. Raimondo

and E. R. Almendra, “The influence of microstructure on the performance of white

porcelain stoneware”, Ceramics International, Vol. 30, pp: 953–963, 2004.

2. E. Rambaldi, L. Esposito, A. Tucci and G. Timellini, “Recycling of polishing

porcelain stoneware residues in ceramic tiles”, Journal of the European Ceramic

Society, Vol. 27, pp: 3509–3515, 2007.

3. I. M. Hutchings, K. Adachi, Y. Xu, E. Sanchez, M. J. Ibanez and M. F. Quereda,

“Analysis and laboratory simulation of an industrial polishing process for porcelain

ceramic tiles”, Journal of the European Ceramic Society, Vol. 25, pp: 3151-3156,

2005.

4. J. Martin-Marquez, J. M. Rincon and M. Romero, “Effect of firing temperature

on sintering of porcelain stoneware tiles”, Ceramics International, Vol. 34, pp:

1867-1873, 2008.

5. M. Dondi, G. Ercolani, G. Guarini, C. Melandri, M. Raimondo, E. R. Almendra

and P. M. Tenorio Cavalcante, “The role of surface microstructure on the resistance

to stains of porcelain stoneware tiles”, Journal of the European Ceramic Society,

Vol. 25, pp: 357–365, 2005.

6. E. Sanchez, M. J. Ibanez, J. Garcıa-Ten, M.F. Quereda, I.M. Hutchings and Y.

M. Xu, “Porcelain tile microstructure: Implications for polished tile properties”,

Journal of the European Ceramic Society, Vol. 26, pp: 2533–2540, 2006.

7. F. Andreola, L. Barbieri, E. Karamanova, I. Lancellotti and M. Pelino,

“Recycling of CRT panel glass as fluxing agent in the porcelain stoneware tile

production”, Ceramics International, Vol. 34, pp: 1289–1295, 2008.

70

8. C. Leonelli, F. Bondioli, P. Veronesi, M. Romagnoli, T. Manfredini, G. C.

Pellacani and V. Cannillo, “Enhancing the mechanical properites of porcelain

stoneware tiles: a microstructural approach”, Journal of the European Ceramic

Society, Vol. 21, pp: 785-793, 2001.

9. M. Romero, J. M. Marquez and J. M. Rincon, “Kinetic of mullite formation from

a porcelain stoneware body for tiles production”, Journal of the European Ceramic

Society, Vol. 26, pp: 1647–1652, 2006.

10. I. M. Hutchings, Y. Xua, E. Sanchez, M. J. Ibanez and M. F. Quereda,

“Porcelain tile microstructure: implications for polishability”, Journal of the

European Ceramic Society, Vol. 26, pp: 1035–1042, 2006.

11. L. Esposito, A. Tucci and D. Naldi, “The reliability of polished porcelain

stoneware tiles”, Journal of the European Ceramic Society, Vol. 25, pp: 1487–1498,

2005.

12. E. Sanchez, J. Garcia-Ten, M. J Ibanez, M. J Orts and V. Cantavella, “Polishing

porcelain tile (Part 1 wear mechanism)”, American Ceramic Society Bulletin, Vol.

81, pp: 50-54, 2002.

13. C. Y. Wang, X. Wei and H. Yuan, “Polishing of ceramic tiles”, Materials and

Manufacturing Processes, Vol. 17, pp: 401-413, 2002.

14. Ö. Güzel and L. Gündüz, “Abrasives used in marble polishing cycles – an

alternative matrix form analyses”, Türkiye III. Mermer Sempozyumu (Mersem

2001), pp: 249-263, 2001.

15. M. D. Casella, J. C. Lorentz, A. Traveria and J. M Tura, “Cracks in Sorel’s

Cement polishing bricks as a result of magnesium oxychloride carbonatation”,

Cement and Concrete Research, Vol. 26, pp: 1199-1202, 1996.

71

16. L. Zongjin and C. K. Chau, “Influence of molar ratios on properties of

magnesium oxychloride cement”, Cement and Concrete Research, Vol. 37, pp:

866–870, 2007.

17. K. P. Maravelaki and G. Moraitou, “Sorel’s Cement mortars decay

susceptibility and effect on pentelic marble”, Cement and Concrete Research, Vol.

29, pp: 1929–1935, 1999.

18. M. S. Özer, A. Göktaş, A. Öztürk and M. Timuçin, “Production and

characterization of Sorel Cement based abrasive bricks for surface polishing of

ceramic tiles”, SERES 2007 International Ceramic and Glaze Symposium, 2007.

19. M. S. Özer, A. Öztürk and M. Timuçin, “Formation and characterization of

Sorel Cements”, International Symposium of Sustainability in Cement and

Concrete, 2006.

20. Türk Standardları Enstitüsü, “TS 1769 Klinkerler- Sorel Çimentosu klinkeri

(MgO)”, Ankara, 1994.

21. W. O. Robinson and W. H. Wingman, “Basic magnesium chlorides”, Journal of

Physical Chemistry, Vol. 13, pp: 673-678, 1909.

22. C. R. Bury and E. R. H. Davies, “System magnesium oxide-magnesium

chloride-water”, Journal of Chemical Society (London), pp: 2008-2015, 1932.

23. C. Harper, “Effect of calcination temperature on the properties of magnesium

oxides for use in magnesium oxychloride cements”, Journal of Applied Chemistry,

Vol. 17, pp: 5-10, 1967.

24. C. A. Sorrell and C. R. Armstrong, “Reactions and equilibria in magnesium

oxychloride cements”, Journal of the American Ceramic Society, Vol. 59, pp: 51-

54, 1976.

72

25. M. C. Ball, “Reactions of compounds occurring in Sorel’s Cement”, Cement

and Concrete Research, Vol. 7, pp: 575-584, 1977.

26. L. Urwongse and C. A. Sorrell, “The system MgO-MgCl2-H2O at 23 °C”,

Journal of the American Ceramic Society, Vol. 63, pp: 501-504, 1980.

27. C. Mazuranic, H. Bilinski and B. Matkovic, “Reaction products in the system

MgCl2-NaOH-H2O”, Journal of the American Ceramic Society, Vol. 65, pp: 523-

26, 1982.

28. H. Bilinski, B. Matkovic, C. Mazuranic and T. B. Zunic, “The formation of

magnesium oxychloride phases in the systems MgO-MgCl2-H2O and NaOH-

MgCl2-H2O.” Journal of the American Ceramic Society, Vol. 67, pp: 266-269,

1984.

29. D. Dehua and Z. Chuanmei, “The formation mechanism of the hydrate phases in

magnesium oxychloride cement”, Cement and Concrete Research, Vol. 29, pp:

1365–1371, 1999.

30. J. Yunsong, “A new type of light magnesium cement foamed material”,

Materials Letters, Vol. 56, pp: 353– 356, 2002.

31. L. A. Skomorovskaya, T. S. Rudenko and E. A. Gabelkova: “Single firing of

ceramic tiles for interior-wall facing”, An Imprint of Springer Verlag, New York

LLC., Vol. 50, No. 7, pp: 304-305, 1993.

32. M. Llusar, C. Rodrigues, J. Labrincha, M. Flores and G. Monrós,

“Reinforcement of single-firing ceramic glazes with the addition of polycrystalline

tetragonal zirconia (3Y–TZP) or zircon”, Journal of European Ceramic Society,

Vol. 22, pp: 639-652, 2002.

73

33. M. Llusar, G. Monrós, C. M. Rodrigues and J. A. Labrincha, “Study of zircon or

zirconia crystals addition in ceramic glazes by impedance spectroscopy”, Ceramic

International, Vol. 31, pp: 181-188, 2005.

34. M. S. Gasser, G. H. A. Morad and H. F. Aly, “Batch kinetics and

thermodynamics of chromium ions removal from waste solutions using synthetic

adsorbents”, Journal of Hazardous Materials, Vol. 142, pp: 118 –129, 2007.

35. D. Dehua, “The mechanism for soluble phosphates to improve the water

resistance of magnesium oxychloride cement”, Cement and Concrete Research,

Vol. 33, pp: 1311 –1317, 2003.

36. P. Torres, H. R. Fernandes, S. Agathopoulos, D. U. Tulyaganov and J. M. F.

Ferreira, “Incorporation of granite cutting sludge in industrial porcelain tile

formulations”, Journal of the European Ceramic Society, Vol. 24, pp: 3177–3185,

2004.

37. P. Torres, R. S. Manjate, S. Quaresma, H. R. Fernandes and J. M. F. Ferreira,

“Development of ceramic floor tile compositions based on quartzite and granite

sludges”, Journal of the European Ceramic Society, Vol. 27, pp: 4649–4655, 2007.

38. E. Rambaldi, W. M. Carty, A. Tucci and L. Esposito, “Using waste glass as a

partial flux substitution and pyroplastic deformation of a porcelain stoneware tile

body”, Ceramics International, Vol. 33, pp: 727–733, 2007.

39. D. Sarı and H. Yavuz, “Quantitative definition of marble gloss”, Türkiye 3.

Mermer Sempozyumu (Mersem 2001), pp: 265-275, 2001.

40. S. Dönmez and Y. D. Sarı, “3-Dimensional surface roughness measuring

system”, Teknoloji, Vol. 7, pp: 41-49, 2004.

74

41. W. Chang, “The effects of surface roughness and contaminants on the dynamic

friction between porcelain tile and vulcanized rubber”, Safety Science, Vol. 40, pp:

577-591, 2002.

42. J. Li, G. Li and Y. Yu, “The influence of compound additive on magnesium

oxychloride cement/urban refuse floor tile”, Construction and Building Materials,

Vol. 22, pp: 521-525, 2008.

43. Greecian Magnesite Co., Products catologue (Coustic MgO).

44. Nedmag Industires Mining & Manufacturing B.V., Product data sheet (MgCl2).

45. Akyol T.A.Ş. Products catologue (Zımpara kumu).

46. Eczacıbaşı Esan, Technical Data Sheet, No: TL.19.PS.70, 2005.

47. Türk Standardları Enstitüsü, “TS 1769 Klinkerler- Sorel Çimentosu Klinkeri

(MgO)”, Ankara, 1994.

48. Türk Standardları Enstitüsü, “TS 699 - Tabii Yapı Taşları-Muayene ve Deney

Metotları”, Ankara, 1987.

49. W. Fan, X. Song, S. Sun and X. Zhao, “Hydrothermal formation and

characterization of magnesium hydroxide chloride hydrate nanowires”, Journal of

Crystal Growth, Vol. 305, pp: 167–174, 2007.

50. L. J. Vandeperre, M. Liska and A. Al-Tabbaa, “Microstructures of reactive

magnesia blends”, Cement and Concrete Composites, Vol. 30, pp: 706-714, 2008.

51. M. Masi, D. Colella, G. Radaelli and L. Bertolini, “Simulation of chloride

penetration in cement-based materials”, Cement and Concrete Research, Vol. 27,

pp: 1591-1601, 1997.

75

52. M. Lanzon and P. A. G. Ruiz, “Lightweight cement mortars: advantages and

inconveniences of expanded perlite and its influence on fresh and hardened state

and durability”, Construction and Building Materials, Vol. 22, pp: 1798-1806, 2008.

53. O. Bernard, F. J. Ulm and E. Lemarchand, “A multiscale micromechanics-

hydration model for the early-age elastic properties of cement-based materials”,

Cement and Concrete Research, Vol. 33, pp: 1293–1309, 2003.

54. G. Sant, C. F. Ferraris and J. Weiss, “Rheological properties of cement pastes: a

discussion of structure formation and mechanical property development”, Cement

and Concrete Research, Vol. 38, pp: 1286-1296, 2008.

55. X. Zhang and J. Han, “The effect of ultra-fine admixtures on the rheological

property of cement paste”, Cement and Concrete Research, Vol. 30, pp: 827-830,

2000.

56. O. Özer and M. S. Kınkoğlu, “Türkiye'de üretilen magnezyaların Sorel

Çimentosu parametrelerinin incelenmesi”, 4. Endüstriyel Hammaddeler

Sempozyumu, 2001.

57. R. Gennaro, A. Langella, M. D. Amore, M. Dondi, A. Colella, P. Cappelletti

and M. Gennaro, “Use of zeolite-rich rocks and waste materials for the production

of structural lightweight concretes”, Applied Clay Science, Vol. 41, pp: 61-72,

2008.

58. S. Tsivilisa, J. Tsantilasa, G. Kakalia, E. Chaniotakisb and A. Sakellariou, “The

permeability of Portland limestone cement concrete”, Cement and Concrete

Research, Vol. 33, pp: 1465–1471, 2003.

59. S.P. Pandey, A. K. Singh, R. L. Sharma and A. K. Tiwari, “Studies on high-

performance blended/multiblended cements and their durability characteristics”,

Cement and Concrete Research, Vol. 33, pp: 1433–1436, 2003.

76

60. N. Bouzoubaa, M. H. Zhang and V. M. Malhotra, “Laboratory-produced high-

volume fly ash blended cements compressive strength and resistance to the

chloride-ion penetration of concrete”, Cement and Concrete Research, Vol. 30, pp:

1037-1046, 2000.

61. L. Basheer, P. A. M. Basheer and A. E. Long, “Influence of coarse aggregate on

the permeation, durability and the microstructure characteristics of ordinary

portland cement concrete”, Construction and Building Materials, Vol. 19, pp: 682–

690, 2005.

62. J. W. Bullard and E. J. Garboczi, “A model investigation of the influence of

particle shape on portland cement hydration”, Cement and Concrete Research,

Vol. 36, pp: 1007-1015, 2006.

63. N. Bouzoubaa, M. H. Zhang and V. M. Malhotra, “Mechanical properties and

durability of concrete made with high-volume fly ash blended cements using a

coarse fly ash”, Cement and Concrete Research, Vol. 31, pp: 1393-1402, 2001.

64. J. M. Skibo, T. C. Butts and M. B. Schiffer, “Ceramic surface treatment and

abrasion resistance: an experimental study”, Journal of Archaeological Science,

Vol. 24, pp: 311-317, 1997.

65. C. Carde and R. Frarqois, “Modelling the loss of strength and porosity increase

due to the leaching of cement pastes”, Cement and Concrete Composites, Vol. 21,

pp: 181-188, 1999.

66. Z. Lafhaj, M. Goueygou, A. Djerbi and M. Kaczmarek, “Correlation between

porosity, permeability and ultrasonic parameters of mortar with variable

water/cement ratio and water content”, Cement and Concrete Research, Vol. 36, pp:

625-633, 2006.

77

67. C. Freidin, “Alkali-activated cement based on natural SiO2-containing material

Part I. strength, hydration, microstructure and durability”, Cement and Concrete

Research, Vol. 33, pp: 1417-1422, 2003.

Production and Characterization of Magnesium Oxychloride ... · production and characterization of...

Documents

Transcript of Production and Characterization of Magnesium Oxychloride ... · production and characterization of...